diff --git a/datasets.ipynb b/datasets.ipynb new file mode 100644 index 0000000..cb8a4cb --- /dev/null +++ b/datasets.ipynb @@ -0,0 +1,4187 @@ +{ + "cells": [ + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "# Datasets Guide\n", + "\n", + "\n", + "- The `DatasetLoader` class provides a simple interface to access example multi-omics datasets included in this package. \n", + "- Each dataset is loaded as a collection of **pandas DataFrames**, with table names as keys and the corresponding data as values. \n", + "- Users can explore the structure of any dataset via the `.shape` property, which returns a mapping from table name to `(rows, columns)`. \n", + "- Three datasets are available out-of-the-box:\n", + "\n", + " 1. **Example1**:\n", + " - Synthetic dataset designed for testing and demonstration \n", + " - Contains small DataFrames: `X1`, `X2`, `Y`, `clinical_data` \n", + " - Useful for quick checks of package functionality\n", + "\n", + " 2. **monet**: \n", + " - Multi-omics benchmark dataset from the **Multi-Omics NETwork Analysis Workshop (MONET)**. \n", + " - Includes multiple DataFrames: `gene_data`, `mirna_data`, `phenotype`, `rppa_data`, `clinical_data` \n", + " - Workshop details: \n", + "\n", + " 3. **brca** :\n", + " - Breast cancer cohort from TCGA (BRCA project) \n", + " - Provides comprehensive omics DataFrames: `rna`, `mirna`, `meth`, `pam50`, `clinical` \n", + " - Full dataset description available at: \n", + "\n" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [ + { + "data": { + "text/plain": [ + "'example1 shapes'" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/plain": [ + "' - X1: 358 x 500'" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/plain": [ + "' - X2: 358 x 100'" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/plain": [ + "' - Y: 358 x 1'" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/plain": [ + "' - clinical_data: 358 x 6'" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/plain": [ + "'monet shapes'" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/plain": [ + "' - gene_data: 107 x 5039'" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/plain": [ + "' - mirna_data: 107 x 789'" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/plain": [ + "' - phenotype: 106 x 1'" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/plain": [ + "' - rppa_data: 107 x 175'" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/plain": [ + "' - clinical_data: 107 x 5'" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/plain": [ + "'brca shapes'" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/plain": [ + "' - mirna: 769 x 503'" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/plain": [ + "' - pam50: 769 x 1'" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/plain": [ + "' - clinical: 769 x 118'" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/plain": [ + "' - rna: 769 x 6000'" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/plain": [ + "' - meth: 769 x 6000'" + ] + }, + "metadata": {}, + "output_type": "display_data" + } + ], + "source": [ + "from bioneuralnet.datasets import DatasetLoader\n", + "import pandas as pd\n", + "\n", + "for name in [\"example1\", \"monet\", \"brca\"]:\n", + " ds = DatasetLoader(name)\n", + " display(f\"{name} shapes:\")\n", + " for tbl, (rows, cols) in ds.shape.items():\n", + " display(f\" - {tbl}: {rows} x {cols}\")" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "### Example 1: Synthetic dataset" + ] + }, + { + "cell_type": "code", + "execution_count": 4, + "metadata": {}, + "outputs": [ + { + "data": { + "text/html": [ + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
Gene_1Gene_2Gene_3Gene_4Gene_5
Samp_122.48570140.35372031.02574520.84720626.697293
Samp_237.05885034.05223333.48702023.53146126.754628
Samp_320.53076731.66962335.18956720.95254425.018826
Samp_433.18688838.48088018.89709731.82330034.049383
Samp_528.96198141.06049428.49495618.37449530.815238
..................
Samp_35424.52065228.59540931.29966632.09537933.659730
Samp_35531.25278928.98808729.57419531.18928832.098841
Samp_35624.89482625.94488730.85264126.70515830.102546
Samp_35717.03433738.57470525.09520137.06244235.417758
Samp_35820.83916727.09978831.03845319.41085931.818995
\n", + "

358 rows × 5 columns

\n", + "
" + ], + "text/plain": [ + " Gene_1 Gene_2 Gene_3 Gene_4 Gene_5\n", + "Samp_1 22.485701 40.353720 31.025745 20.847206 26.697293\n", + "Samp_2 37.058850 34.052233 33.487020 23.531461 26.754628\n", + "Samp_3 20.530767 31.669623 35.189567 20.952544 25.018826\n", + "Samp_4 33.186888 38.480880 18.897097 31.823300 34.049383\n", + "Samp_5 28.961981 41.060494 28.494956 18.374495 30.815238\n", + "... ... ... ... ... ...\n", + "Samp_354 24.520652 28.595409 31.299666 32.095379 33.659730\n", + "Samp_355 31.252789 28.988087 29.574195 31.189288 32.098841\n", + "Samp_356 24.894826 25.944887 30.852641 26.705158 30.102546\n", + "Samp_357 17.034337 38.574705 25.095201 37.062442 35.417758\n", + "Samp_358 20.839167 27.099788 31.038453 19.410859 31.818995\n", + "\n", + "[358 rows x 5 columns]" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
Mir_1Mir_2Mir_3Mir_4Mir_5
Samp_115.22391317.54582615.78471914.89198310.348205
Samp_216.30696516.67283013.36152914.48854912.660905
Samp_316.54511916.73500514.61747217.84526713.822790
Samp_413.98689916.20743216.29307817.72528612.300565
Samp_516.33833217.39386916.39792515.85372513.387675
..................
Samp_35415.06506516.07983014.63561617.01384511.612843
Samp_35515.99757615.44895115.35556616.50175211.701778
Samp_35615.20686214.39537816.21800116.04495513.650741
Samp_35714.47412915.48286315.51254915.13661314.531277
Samp_35815.09418816.04730415.29887117.02266515.046043
\n", + "

358 rows × 5 columns

\n", + "
" + ], + "text/plain": [ + " Mir_1 Mir_2 Mir_3 Mir_4 Mir_5\n", + "Samp_1 15.223913 17.545826 15.784719 14.891983 10.348205\n", + "Samp_2 16.306965 16.672830 13.361529 14.488549 12.660905\n", + "Samp_3 16.545119 16.735005 14.617472 17.845267 13.822790\n", + "Samp_4 13.986899 16.207432 16.293078 17.725286 12.300565\n", + "Samp_5 16.338332 17.393869 16.397925 15.853725 13.387675\n", + "... ... ... ... ... ...\n", + "Samp_354 15.065065 16.079830 14.635616 17.013845 11.612843\n", + "Samp_355 15.997576 15.448951 15.355566 16.501752 11.701778\n", + "Samp_356 15.206862 14.395378 16.218001 16.044955 13.650741\n", + "Samp_357 14.474129 15.482863 15.512549 15.136613 14.531277\n", + "Samp_358 15.094188 16.047304 15.298871 17.022665 15.046043\n", + "\n", + "[358 rows x 5 columns]" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
phenotype
Samp_1235.067423
Samp_2253.544991
Samp_3234.204994
Samp_4281.035429
Samp_5245.447781
......
Samp_354236.120451
Samp_355222.572359
Samp_356268.472285
Samp_357235.808167
Samp_358213.886123
\n", + "

358 rows × 1 columns

\n", + "
" + ], + "text/plain": [ + " phenotype\n", + "Samp_1 235.067423\n", + "Samp_2 253.544991\n", + "Samp_3 234.204994\n", + "Samp_4 281.035429\n", + "Samp_5 245.447781\n", + "... ...\n", + "Samp_354 236.120451\n", + "Samp_355 222.572359\n", + "Samp_356 268.472285\n", + "Samp_357 235.808167\n", + "Samp_358 213.886123\n", + "\n", + "[358 rows x 1 columns]" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
AgeGenderBMIChronic_BronchitisEmphysema
PatientID
Samp_178031.211
Samp_268119.210
Samp_354119.301
Samp_447136.200
Samp_560126.201
..................
Samp_35471023.010
Samp_35562125.501
Samp_35661021.110
Samp_35764037.600
Samp_35861131.310
\n", + "

358 rows × 5 columns

\n", + "
" + ], + "text/plain": [ + " Age Gender BMI Chronic_Bronchitis Emphysema\n", + "PatientID \n", + "Samp_1 78 0 31.2 1 1\n", + "Samp_2 68 1 19.2 1 0\n", + "Samp_3 54 1 19.3 0 1\n", + "Samp_4 47 1 36.2 0 0\n", + "Samp_5 60 1 26.2 0 1\n", + "... ... ... ... ... ...\n", + "Samp_354 71 0 23.0 1 0\n", + "Samp_355 62 1 25.5 0 1\n", + "Samp_356 61 0 21.1 1 0\n", + "Samp_357 64 0 37.6 0 0\n", + "Samp_358 61 1 31.3 1 0\n", + "\n", + "[358 rows x 5 columns]" + ] + }, + "metadata": {}, + "output_type": "display_data" + } + ], + "source": [ + "from bioneuralnet.datasets import DatasetLoader\n", + "\n", + "Example = DatasetLoader(\"example1\")\n", + "omics1 = Example.data[\"X1\"]\n", + "omics2= Example.data[\"X2\"]\n", + "phenotype = Example.data[\"Y\"]\n", + "clinical = Example.data[\"clinical_data\"]\n", + "\n", + "display(omics1.iloc[:, :5])\n", + "display(omics2.iloc[:, :5])\n", + "display(phenotype.iloc[:, :5])\n", + "display(clinical.iloc[:, :5])" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "### Monet: Set from the **Multi-Omics NETwork Analysis Workshop (MONET)**, Univ. of Colorado Anschutz " + ] + }, + { + "cell_type": "code", + "execution_count": 5, + "metadata": {}, + "outputs": [ + { + "data": { + "text/html": [ + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
A2ML1AACSLAADACAADATAATK
00.4666710.0748450.990309-0.4108731.897562
1-0.524465-0.146727-0.735206-0.6284560.170962
2-0.029879-0.626509-0.735206-0.6778920.020060
30.674895-0.626509-0.330409-0.662162-0.911966
4-0.110607-0.626509-0.735206-0.8485420.042645
..................
1020.9996001.343979-0.7352061.7426740.253103
103-0.919337-0.626509-0.735206-1.380461-0.899019
104-0.606702-0.6265090.497658-0.717505-0.625122
1051.9113460.021380-0.1666240.7855420.344953
106-0.444363-0.111055-0.735206-1.395805-0.685468
\n", + "

107 rows × 5 columns

\n", + "
" + ], + "text/plain": [ + " A2ML1 AACSL AADAC AADAT AATK\n", + "0 0.466671 0.074845 0.990309 -0.410873 1.897562\n", + "1 -0.524465 -0.146727 -0.735206 -0.628456 0.170962\n", + "2 -0.029879 -0.626509 -0.735206 -0.677892 0.020060\n", + "3 0.674895 -0.626509 -0.330409 -0.662162 -0.911966\n", + "4 -0.110607 -0.626509 -0.735206 -0.848542 0.042645\n", + ".. ... ... ... ... ...\n", + "102 0.999600 1.343979 -0.735206 1.742674 0.253103\n", + "103 -0.919337 -0.626509 -0.735206 -1.380461 -0.899019\n", + "104 -0.606702 -0.626509 0.497658 -0.717505 -0.625122\n", + "105 1.911346 0.021380 -0.166624 0.785542 0.344953\n", + "106 -0.444363 -0.111055 -0.735206 -1.395805 -0.685468\n", + "\n", + "[107 rows x 5 columns]" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
hsa-let-7a-1hsa-let-7a-2hsa-let-7a-3hsa-let-7bhsa-let-7c
0-0.832527-0.851616-0.837155-1.079659-0.181270
10.2291550.2496960.2342950.859289-0.057729
20.4142680.4170230.4089130.6350591.195203
3-0.855214-0.869152-0.862713-1.955447-0.572552
41.3653101.3562521.3517501.259095-0.316760
..................
102-1.402001-1.401125-1.349961-1.5343861.231456
1032.5512772.5470462.5631911.0547690.981436
1040.1821380.1887300.191094-1.0606150.345907
1050.2894890.2924700.297778-0.1978501.040270
1060.7910680.8006800.8036580.5895980.937994
\n", + "

107 rows × 5 columns

\n", + "
" + ], + "text/plain": [ + " hsa-let-7a-1 hsa-let-7a-2 hsa-let-7a-3 hsa-let-7b hsa-let-7c\n", + "0 -0.832527 -0.851616 -0.837155 -1.079659 -0.181270\n", + "1 0.229155 0.249696 0.234295 0.859289 -0.057729\n", + "2 0.414268 0.417023 0.408913 0.635059 1.195203\n", + "3 -0.855214 -0.869152 -0.862713 -1.955447 -0.572552\n", + "4 1.365310 1.356252 1.351750 1.259095 -0.316760\n", + ".. ... ... ... ... ...\n", + "102 -1.402001 -1.401125 -1.349961 -1.534386 1.231456\n", + "103 2.551277 2.547046 2.563191 1.054769 0.981436\n", + "104 0.182138 0.188730 0.191094 -1.060615 0.345907\n", + "105 0.289489 0.292470 0.297778 -0.197850 1.040270\n", + "106 0.791068 0.800680 0.803658 0.589598 0.937994\n", + "\n", + "[107 rows x 5 columns]" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
0
01
10
20
30
40
......
1010
1021
1030
1040
1050
\n", + "

106 rows × 1 columns

\n", + "
" + ], + "text/plain": [ + " 0\n", + "0 1\n", + "1 0\n", + "2 0\n", + "3 0\n", + "4 0\n", + ".. ..\n", + "101 0\n", + "102 1\n", + "103 0\n", + "104 0\n", + "105 0\n", + "\n", + "[106 rows x 1 columns]" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
YWHAEYWHAZEIF4EBP1TP53BP1ARAF
0-0.3579980.099812-1.067285-0.412211-0.357998
1-0.055031-0.5174450.032633-0.743096-0.055031
2-0.137863-0.5596900.302764-0.968388-0.137863
3-0.170726-0.028206-0.3414610.282581-0.170726
4-1.430765-0.138087-0.545894-0.616864-1.430765
..................
102-0.708685-0.7788131.623365-0.090612-0.708685
1030.261442-0.407563-0.567735-0.1869190.261442
1041.3508661.461061-1.159541-1.6748741.350866
1050.179510-0.300029-1.048938-0.6216800.179510
106-1.099437-0.818285-0.855339-1.313032-1.099437
\n", + "

107 rows × 5 columns

\n", + "
" + ], + "text/plain": [ + " YWHAE YWHAZ EIF4EBP1 TP53BP1 ARAF\n", + "0 -0.357998 0.099812 -1.067285 -0.412211 -0.357998\n", + "1 -0.055031 -0.517445 0.032633 -0.743096 -0.055031\n", + "2 -0.137863 -0.559690 0.302764 -0.968388 -0.137863\n", + "3 -0.170726 -0.028206 -0.341461 0.282581 -0.170726\n", + "4 -1.430765 -0.138087 -0.545894 -0.616864 -1.430765\n", + ".. ... ... ... ... ...\n", + "102 -0.708685 -0.778813 1.623365 -0.090612 -0.708685\n", + "103 0.261442 -0.407563 -0.567735 -0.186919 0.261442\n", + "104 1.350866 1.461061 -1.159541 -1.674874 1.350866\n", + "105 0.179510 -0.300029 -1.048938 -0.621680 0.179510\n", + "106 -1.099437 -0.818285 -0.855339 -1.313032 -1.099437\n", + "\n", + "[107 rows x 5 columns]" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
overall_survivalstatusyears_to_birthraceradiation_therapy
03015037blackorafricanamericanyes
12348173whiteyes
23011041asianyes
33283067whiteno
41873042whiteno
..................
1022329063whiteyes
1031004174whiteyes
104984046whiteyes
105867044whiteno
1061133053whiteyes
\n", + "

107 rows × 5 columns

\n", + "
" + ], + "text/plain": [ + " overall_survival status years_to_birth race \\\n", + "0 3015 0 37 blackorafricanamerican \n", + "1 2348 1 73 white \n", + "2 3011 0 41 asian \n", + "3 3283 0 67 white \n", + "4 1873 0 42 white \n", + ".. ... ... ... ... \n", + "102 2329 0 63 white \n", + "103 1004 1 74 white \n", + "104 984 0 46 white \n", + "105 867 0 44 white \n", + "106 1133 0 53 white \n", + "\n", + " radiation_therapy \n", + "0 yes \n", + "1 yes \n", + "2 yes \n", + "3 no \n", + "4 no \n", + ".. ... \n", + "102 yes \n", + "103 yes \n", + "104 yes \n", + "105 no \n", + "106 yes \n", + "\n", + "[107 rows x 5 columns]" + ] + }, + "metadata": {}, + "output_type": "display_data" + } + ], + "source": [ + "from bioneuralnet.datasets import DatasetLoader\n", + "\n", + "monet = DatasetLoader(\"monet\")\n", + "gene = monet.data[\"gene_data\"]\n", + "mirna = monet.data[\"mirna_data\"]\n", + "phenotype = monet.data[\"phenotype\"]\n", + "rppa = monet.data[\"rppa_data\"]\n", + "clinical = monet.data[\"clinical_data\"]\n", + "\n", + "display(gene.iloc[:, :5])\n", + "display(mirna.iloc[:, :5])\n", + "display(phenotype.iloc[:, :5])\n", + "display(rppa.iloc[:, :5])\n", + "display(clinical.iloc[:, :5])" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "### BRCA: Breast cancer cohort dataset." + ] + }, + { + "cell_type": "code", + "execution_count": 6, + "metadata": {}, + "outputs": [ + { + "data": { + "text/html": [ + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
ESR1_2099FOXA1_3169MLPH_79083AGR3_155465TBC1D9_23158
patient
TCGA-3C-AAAU11.75570612.41160812.56677512.04972914.173402
TCGA-3C-AALI6.09835812.56259614.10126312.43169111.295692
TCGA-3C-AALJ12.86927012.17371712.31543511.49609812.314665
TCGA-3C-AALK11.27921112.84393913.37929110.15357112.610953
TCGA-4H-AAAK12.43000812.73122912.58092010.25367212.353710
..................
TCGA-WT-AB4412.15442112.58394915.22331211.01516411.632502
TCGA-XX-A89911.41547612.00154713.0672129.70433912.868580
TCGA-XX-A89A11.28757611.98877112.76982510.19002511.919563
TCGA-Z7-A8R511.68885211.54486112.52218610.55692412.289650
TCGA-Z7-A8R612.65058612.97678712.83372411.66858712.534803
\n", + "

769 rows × 5 columns

\n", + "
" + ], + "text/plain": [ + " ESR1_2099 FOXA1_3169 MLPH_79083 AGR3_155465 TBC1D9_23158\n", + "patient \n", + "TCGA-3C-AAAU 11.755706 12.411608 12.566775 12.049729 14.173402\n", + "TCGA-3C-AALI 6.098358 12.562596 14.101263 12.431691 11.295692\n", + "TCGA-3C-AALJ 12.869270 12.173717 12.315435 11.496098 12.314665\n", + "TCGA-3C-AALK 11.279211 12.843939 13.379291 10.153571 12.610953\n", + "TCGA-4H-AAAK 12.430008 12.731229 12.580920 10.253672 12.353710\n", + "... ... ... ... ... ...\n", + "TCGA-WT-AB44 12.154421 12.583949 15.223312 11.015164 11.632502\n", + "TCGA-XX-A899 11.415476 12.001547 13.067212 9.704339 12.868580\n", + "TCGA-XX-A89A 11.287576 11.988771 12.769825 10.190025 11.919563\n", + "TCGA-Z7-A8R5 11.688852 11.544861 12.522186 10.556924 12.289650\n", + "TCGA-Z7-A8R6 12.650586 12.976787 12.833724 11.668587 12.534803\n", + "\n", + "[769 rows x 5 columns]" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
hsa_let_7a_1hsa_let_7a_2hsa_let_7a_3hsa_let_7bhsa_let_7c
patient
TCGA-3C-AAAU13.12976514.11793313.14771414.5951358.414890
TCGA-3C-AALI12.91806913.92230012.91319414.5126579.646536
TCGA-3C-AALJ13.01203314.01000213.02848313.4196129.312455
TCGA-3C-AALK13.14469714.14172113.15128114.66719611.511431
TCGA-4H-AAAK13.41168414.41351813.42048114.43854811.693927
..................
TCGA-WT-AB4413.37571514.36667113.36982714.51402411.926315
TCGA-XX-A89914.03615515.03634114.04331314.33950312.361761
TCGA-XX-A89A13.67956914.68485513.69146314.19820712.684212
TCGA-Z7-A8R512.96208813.96635012.98489714.32066011.980246
TCGA-Z7-A8R613.34971114.34990113.37844714.11721210.378041
\n", + "

769 rows × 5 columns

\n", + "
" + ], + "text/plain": [ + " hsa_let_7a_1 hsa_let_7a_2 hsa_let_7a_3 hsa_let_7b hsa_let_7c\n", + "patient \n", + "TCGA-3C-AAAU 13.129765 14.117933 13.147714 14.595135 8.414890\n", + "TCGA-3C-AALI 12.918069 13.922300 12.913194 14.512657 9.646536\n", + "TCGA-3C-AALJ 13.012033 14.010002 13.028483 13.419612 9.312455\n", + "TCGA-3C-AALK 13.144697 14.141721 13.151281 14.667196 11.511431\n", + "TCGA-4H-AAAK 13.411684 14.413518 13.420481 14.438548 11.693927\n", + "... ... ... ... ... ...\n", + "TCGA-WT-AB44 13.375715 14.366671 13.369827 14.514024 11.926315\n", + "TCGA-XX-A899 14.036155 15.036341 14.043313 14.339503 12.361761\n", + "TCGA-XX-A89A 13.679569 14.684855 13.691463 14.198207 12.684212\n", + "TCGA-Z7-A8R5 12.962088 13.966350 12.984897 14.320660 11.980246\n", + "TCGA-Z7-A8R6 13.349711 14.349901 13.378447 14.117212 10.378041\n", + "\n", + "[769 rows x 5 columns]" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
SFT2D2IL17RAMIR128_1FOXA1LOC145837
patient
TCGA-3C-AAAU-2.196646-0.0497423.355022-3.934344-1.801595
TCGA-3C-AALI-2.436039-0.2170062.652026-3.995267-1.512691
TCGA-3C-AALJ-2.390041-0.3601802.564778-3.917724-0.701434
TCGA-3C-AALK-2.469813-0.1077912.718057-4.100320-0.756467
TCGA-4H-AAAK-2.501687-0.0917743.086157-3.628072-0.090305
..................
TCGA-WT-AB44-2.358699-0.0928633.138854-3.864208-0.446164
TCGA-XX-A899-2.633115-0.1926983.330302-3.4984190.144114
TCGA-XX-A89A-2.602103-0.2877182.287165-3.720622-0.236061
TCGA-Z7-A8R5-2.572044-0.1467913.000648-3.3356910.693710
TCGA-Z7-A8R6-2.357329-0.1407414.496094-4.857515-1.210885
\n", + "

769 rows × 5 columns

\n", + "
" + ], + "text/plain": [ + " SFT2D2 IL17RA MIR128_1 FOXA1 LOC145837\n", + "patient \n", + "TCGA-3C-AAAU -2.196646 -0.049742 3.355022 -3.934344 -1.801595\n", + "TCGA-3C-AALI -2.436039 -0.217006 2.652026 -3.995267 -1.512691\n", + "TCGA-3C-AALJ -2.390041 -0.360180 2.564778 -3.917724 -0.701434\n", + "TCGA-3C-AALK -2.469813 -0.107791 2.718057 -4.100320 -0.756467\n", + "TCGA-4H-AAAK -2.501687 -0.091774 3.086157 -3.628072 -0.090305\n", + "... ... ... ... ... ...\n", + "TCGA-WT-AB44 -2.358699 -0.092863 3.138854 -3.864208 -0.446164\n", + "TCGA-XX-A899 -2.633115 -0.192698 3.330302 -3.498419 0.144114\n", + "TCGA-XX-A89A -2.602103 -0.287718 2.287165 -3.720622 -0.236061\n", + "TCGA-Z7-A8R5 -2.572044 -0.146791 3.000648 -3.335691 0.693710\n", + "TCGA-Z7-A8R6 -2.357329 -0.140741 4.496094 -4.857515 -1.210885\n", + "\n", + "[769 rows x 5 columns]" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
pam50
patient
TCGA-3C-AAAU3
TCGA-3C-AALI2
TCGA-3C-AALJ4
TCGA-3C-AALK3
TCGA-4H-AAAK3
......
TCGA-WT-AB443
TCGA-XX-A8993
TCGA-XX-A89A3
TCGA-Z7-A8R53
TCGA-Z7-A8R64
\n", + "

769 rows × 1 columns

\n", + "
" + ], + "text/plain": [ + " pam50\n", + "patient \n", + "TCGA-3C-AAAU 3\n", + "TCGA-3C-AALI 2\n", + "TCGA-3C-AALJ 4\n", + "TCGA-3C-AALK 3\n", + "TCGA-4H-AAAK 3\n", + "... ...\n", + "TCGA-WT-AB44 3\n", + "TCGA-XX-A899 3\n", + "TCGA-XX-A89A 3\n", + "TCGA-Z7-A8R5 3\n", + "TCGA-Z7-A8R6 4\n", + "\n", + "[769 rows x 1 columns]" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
synchronous_malignancyajcc_pathologic_stagedays_to_diagnosislateralitycreated_datetime
patient
TCGA-3C-AAAUNoStage X0.0LeftNaN
TCGA-3C-AALINoStage IIB0.0RightNaN
TCGA-3C-AALJNoStage IIB0.0RightNaN
TCGA-3C-AALKNoStage IA0.0RightNaN
TCGA-4H-AAAKNoStage IIIA0.0LeftNaN
..................
TCGA-WT-AB44NoStage IA0.0LeftNaN
TCGA-XX-A899NoStage IIIA0.0RightNaN
TCGA-XX-A89ANoStage IIB0.0LeftNaN
TCGA-Z7-A8R5NoStage IIIA0.0LeftNaN
TCGA-Z7-A8R6NoStage I0.0LeftNaN
\n", + "

769 rows × 5 columns

\n", + "
" + ], + "text/plain": [ + " synchronous_malignancy ajcc_pathologic_stage days_to_diagnosis \\\n", + "patient \n", + "TCGA-3C-AAAU No Stage X 0.0 \n", + "TCGA-3C-AALI No Stage IIB 0.0 \n", + "TCGA-3C-AALJ No Stage IIB 0.0 \n", + "TCGA-3C-AALK No Stage IA 0.0 \n", + "TCGA-4H-AAAK No Stage IIIA 0.0 \n", + "... ... ... ... \n", + "TCGA-WT-AB44 No Stage IA 0.0 \n", + "TCGA-XX-A899 No Stage IIIA 0.0 \n", + "TCGA-XX-A89A No Stage IIB 0.0 \n", + "TCGA-Z7-A8R5 No Stage IIIA 0.0 \n", + "TCGA-Z7-A8R6 No Stage I 0.0 \n", + "\n", + " laterality created_datetime \n", + "patient \n", + "TCGA-3C-AAAU Left NaN \n", + "TCGA-3C-AALI Right NaN \n", + "TCGA-3C-AALJ Right NaN \n", + "TCGA-3C-AALK Right NaN \n", + "TCGA-4H-AAAK Left NaN \n", + "... ... ... \n", + "TCGA-WT-AB44 Left NaN \n", + "TCGA-XX-A899 Right NaN \n", + "TCGA-XX-A89A Left NaN \n", + "TCGA-Z7-A8R5 Left NaN \n", + "TCGA-Z7-A8R6 Left NaN \n", + "\n", + "[769 rows x 5 columns]" + ] + }, + "metadata": {}, + "output_type": "display_data" + } + ], + "source": [ + "from bioneuralnet.datasets import DatasetLoader\n", + "\n", + "brca = DatasetLoader(\"brca\")\n", + "rna = brca.data[\"rna\"]\n", + "mirna = brca.data[\"mirna\"]\n", + "meth = brca.data[\"meth\"]\n", + "pam50 = brca.data[\"pam50\"]\n", + "clinical = brca.data[\"clinical\"]\n", + "\n", + "display(rna.iloc[:, :5])\n", + "display(mirna.iloc[:, :5])\n", + "display(meth.iloc[:, :5])\n", + "display(pam50.iloc[:, :5])\n", + "display(clinical.iloc[:, :5])" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "## Generating an Omics Network\n", + "At its core, BioNeuralNet leverages Graph Neural Networks to power downstream applications via learned embeddings. It supports a range of standard graph construction techniques for omics and other biological entities:\n", + "\n", + "- **Cosine similarity / RBF kernel graphs** (`gen_similarity_graph`) \n", + "- **Pearson / Spearman correlation graphs** (`gen_correlation_graph`) \n", + "- **Soft-threshold (WGCNA-style) graphs** (`gen_threshold_graph`) \n", + "- **Gaussian k-NN graphs** (`gen_gaussian_knn_graph`) \n", + "- **Mutual information graphs** (`gen_mutual_info_graph`) \n", + "- **Graphical Lasso (sparse inverse covariance) graphs** (`gen_lasso_graph`) \n", + "- **Minimum Spanning Tree (MST) graphs** (`gen_mst_graph`) \n", + "- **Shared Nearest Neighbor (SNN) graphs** (`gen_snn_graph`)\n", + "\n", + "For more details on all of these utilities, see the [utils documentation](https://bioneuralnet.readthedocs.io/en/latest/utils.html). \n" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "## Example: Sparse Multiple Canonical Correlation Network Analysis (SmCCNet 2.0)\n", + "\n", + "In this section, we’ll construct phenotype-specific multi-omics networks using **SmCCNet 2.0**, an R-based algorithm developed by the Kechris Lab at the University of Colorado Anschutz Medical Campus. BioNeuralNet provides a lightweight Python wrapper in `bioneuralnet.external_tools.smccnet` to simplify its usage (requires R).\n", + "\n", + "For full details and examples, see the [External Tools documentation](https://bioneuralnet.readthedocs.io/en/latest/external_tools/index.html).\n", + "\n", + "**Install & Resources** \n", + "- CRAN: https://cran.r-project.org/web/packages/SmCCNet/ \n", + "- GitHub: https://github.com/KechrisLab/SmCCNet \n", + "- Publication: https://bmcbioinformatics.biomedcentral.com/articles/10.1186/s12859-024-05900-9" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "from bioneuralnet.external_tools import SmCCNet\n", + "\n", + "smccnet = SmCCNet(\n", + " phenotype_df=phenotype,\n", + " omics_dfs=[omics1, omics2],\n", + " data_types=[\"genes\", \"mirna\"],\n", + " subSampNum=1000,\n", + ")\n", + "global_network, clusters = smccnet.run()" + ] + }, + { + "cell_type": "code", + "execution_count": 22, + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Global network shape: (600, 600)\n", + "Number of SmCCnet clusters: 3\n" + ] + }, + { + "data": { + "text/html": [ + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
Gene_1Gene_2Gene_3Gene_4Gene_5Gene_6Gene_7Gene_8Gene_9Gene_10...Mir_91Mir_92Mir_93Mir_94Mir_95Mir_96Mir_97Mir_98Mir_99Mir_100
Gene_10.0000000.158521000.0392050.2645890.27512500.0099180.041573...000000.0086870000
Gene_20.1585210.000000000.0355080.2206770.24142800.0082980.034195...000000.0057850000
Gene_30.0000000.000000000.0000000.0000000.00000000.0000000.000000...000000.0000000000
Gene_40.0000000.000000000.0000000.0000000.00000000.0000000.000000...000000.0000000000
Gene_50.0392050.035508000.0000000.0574350.06173700.0021780.009461...000000.0023140000
\n", + "

5 rows × 600 columns

\n", + "
" + ], + "text/plain": [ + " Gene_1 Gene_2 Gene_3 Gene_4 Gene_5 Gene_6 Gene_7 \\\n", + "Gene_1 0.000000 0.158521 0 0 0.039205 0.264589 0.275125 \n", + "Gene_2 0.158521 0.000000 0 0 0.035508 0.220677 0.241428 \n", + "Gene_3 0.000000 0.000000 0 0 0.000000 0.000000 0.000000 \n", + "Gene_4 0.000000 0.000000 0 0 0.000000 0.000000 0.000000 \n", + "Gene_5 0.039205 0.035508 0 0 0.000000 0.057435 0.061737 \n", + "\n", + " Gene_8 Gene_9 Gene_10 ... Mir_91 Mir_92 Mir_93 Mir_94 \\\n", + "Gene_1 0 0.009918 0.041573 ... 0 0 0 0 \n", + "Gene_2 0 0.008298 0.034195 ... 0 0 0 0 \n", + "Gene_3 0 0.000000 0.000000 ... 0 0 0 0 \n", + "Gene_4 0 0.000000 0.000000 ... 0 0 0 0 \n", + "Gene_5 0 0.002178 0.009461 ... 0 0 0 0 \n", + "\n", + " Mir_95 Mir_96 Mir_97 Mir_98 Mir_99 Mir_100 \n", + "Gene_1 0 0.008687 0 0 0 0 \n", + "Gene_2 0 0.005785 0 0 0 0 \n", + "Gene_3 0 0.000000 0 0 0 0 \n", + "Gene_4 0 0.000000 0 0 0 0 \n", + "Gene_5 0 0.002314 0 0 0 0 \n", + "\n", + "[5 rows x 600 columns]" + ] + }, + "metadata": {}, + "output_type": "display_data" + } + ], + "source": [ + "print(\"Global network shape:\", global_network.shape)\n", + "print(\"Number of SmCCnet clusters:\", len(clusters))\n", + "display(global_network.head())" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "## Generate Network Embeddings using `GNNEmbedding`\n", + "\n", + "Next, we transform the constructed network into low-dimensional representations using graph neural networks. The `GNNEmbedding` module in BioNeuralNet supports models like GCN and GAT, enabling integration of omics, clinical, and phenotype data into a unified graph framework. These embeddings serve as compact, informative representations of each subject, suitable for downstream tasks such as classification, clustering, and visualization." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "from bioneuralnet.network_embedding import GNNEmbedding\n", + "\n", + "merged_omics = pd.concat([omics1, omics2], axis=1)\n", + "\n", + "gnn = GNNEmbedding(\n", + " adjacency_matrix=global_network,\n", + " omics_data=merged_omics,\n", + " phenotype_data=phenotype,\n", + " clinical_data=clinical,\n", + " phenotype_col=\"phenotype\",\n", + " tune=True,\n", + " gpu=True,\n", + ")\n", + "gnn.fit()\n", + "embeddings = gnn.embed(as_df=True)" + ] + }, + { + "cell_type": "code", + "execution_count": 21, + "metadata": {}, + "outputs": [ + { + "data": { + "text/html": [ + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
Embed_1Embed_2Embed_3Embed_4Embed_5Embed_6Embed_7Embed_8Embed_9Embed_10...Embed_503Embed_504Embed_505Embed_506Embed_507Embed_508Embed_509Embed_510Embed_511Embed_512
Gene_1-0.001546-0.0000230.985133-0.0003150.680049-0.005195-0.0004610.0647640.021396-0.004405...-0.0010800.7407670.5902360.060455-0.0000610.8803900.0611951.0219390.011463-0.000211
Gene_2-0.000053-0.0000300.409291-0.0003230.258388-0.000368-0.0000410.161861-0.000133-0.000267...-0.0005860.2855040.2357210.173551-0.0000460.3091780.2293260.337168-0.0000940.151875
Gene_30.064611-0.0000320.0906430.0041600.0493450.1268810.0090000.1319680.0120190.032301...0.0199070.0266910.0384440.077817-0.0000210.0229130.0579430.0251250.0120540.074429
Gene_40.059100-0.0000320.0840780.0761440.1027420.1742390.0122250.0750580.0083460.057449...0.0450180.0990480.0770680.125377-0.0000250.1071490.1026790.0535420.0334490.113692
Gene_50.061504-0.0000320.1074600.0615940.1036720.1272940.0079350.1064200.0120460.021869...0.0413660.0741790.0605060.118548-0.0000260.0700810.1169830.0516940.0106140.087940
\n", + "

5 rows × 512 columns

\n", + "
" + ], + "text/plain": [ + " Embed_1 Embed_2 Embed_3 Embed_4 Embed_5 Embed_6 Embed_7 \\\n", + "Gene_1 -0.001546 -0.000023 0.985133 -0.000315 0.680049 -0.005195 -0.000461 \n", + "Gene_2 -0.000053 -0.000030 0.409291 -0.000323 0.258388 -0.000368 -0.000041 \n", + "Gene_3 0.064611 -0.000032 0.090643 0.004160 0.049345 0.126881 0.009000 \n", + "Gene_4 0.059100 -0.000032 0.084078 0.076144 0.102742 0.174239 0.012225 \n", + "Gene_5 0.061504 -0.000032 0.107460 0.061594 0.103672 0.127294 0.007935 \n", + "\n", + " Embed_8 Embed_9 Embed_10 ... Embed_503 Embed_504 Embed_505 \\\n", + "Gene_1 0.064764 0.021396 -0.004405 ... -0.001080 0.740767 0.590236 \n", + "Gene_2 0.161861 -0.000133 -0.000267 ... -0.000586 0.285504 0.235721 \n", + "Gene_3 0.131968 0.012019 0.032301 ... 0.019907 0.026691 0.038444 \n", + "Gene_4 0.075058 0.008346 0.057449 ... 0.045018 0.099048 0.077068 \n", + "Gene_5 0.106420 0.012046 0.021869 ... 0.041366 0.074179 0.060506 \n", + "\n", + " Embed_506 Embed_507 Embed_508 Embed_509 Embed_510 Embed_511 \\\n", + "Gene_1 0.060455 -0.000061 0.880390 0.061195 1.021939 0.011463 \n", + "Gene_2 0.173551 -0.000046 0.309178 0.229326 0.337168 -0.000094 \n", + "Gene_3 0.077817 -0.000021 0.022913 0.057943 0.025125 0.012054 \n", + "Gene_4 0.125377 -0.000025 0.107149 0.102679 0.053542 0.033449 \n", + "Gene_5 0.118548 -0.000026 0.070081 0.116983 0.051694 0.010614 \n", + "\n", + " Embed_512 \n", + "Gene_1 -0.000211 \n", + "Gene_2 0.151875 \n", + "Gene_3 0.074429 \n", + "Gene_4 0.113692 \n", + "Gene_5 0.087940 \n", + "\n", + "[5 rows x 512 columns]" + ] + }, + "metadata": {}, + "output_type": "display_data" + } + ], + "source": [ + "display(embeddings.head())" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "### Embeddings visualization" + ] + }, + { + "cell_type": "code", + "execution_count": 5, + "metadata": {}, + "outputs": [ + { + "name": "stderr", + "output_type": "stream", + "text": [ + "2025-05-23 10:42:56,099 - bioneuralnet.network_embedding.gnn_embedding - INFO - Preparing node labels.\n", + "2025-05-23 10:42:56,149 - bioneuralnet.network_embedding.gnn_embedding - INFO - Node labels prepared successfully and saved to /tmp/tmpra6n3xlt/labels_600_0523_10_42_56.txt.\n" + ] + }, + { + "data": { + "image/png": "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", + "text/plain": [ + "
" + ] + }, + "metadata": {}, + "output_type": "display_data" + } + ], + "source": [ + "from bioneuralnet.metrics import plot_embeddings\n", + "\n", + "# Using our embeddings instance we get the necessary labels for the graph.\n", + "node_labels = gnn._prepare_node_labels()\n", + "embeddings_array = embeddings.values \n", + "\n", + "embeddings_plot = plot_embeddings(embeddings_array, node_labels)" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "## Embedding Interpretation Notes\n", + "\n", + "We projected the 512‑dimensional node embeddings into 2‑D and observed three clear regions plus one broad cloud.\n", + "The class buckets (after binning the continuous phenotype into four equal‑frequency groups) contain:\n", + "\n", + "| Bucket | Samples |\n", + "| ------ | ------- |\n", + "| 0 | 38 |\n", + "| 1 | 158 |\n", + "| 2 | 141 |\n", + "| 3 | 21 |\n", + "\n", + "## Visual observations\n", + "\n", + "- **Upper‑left cloud**\n", + "\n", + " - Largest and most diffuse region\n", + " - Contains the majority of points\n", + " - Most likely accounting for buckets 1 and 2\n", + "\n", + "- **Right‑hand oval**\n", + "\n", + " - Compact cluster on the far right\n", + " - Roughly forty points\n", + " - Could be bucket 0 or 3\n", + "\n", + "- **Lower cluster**\n", + "\n", + " - Compact group far below the main cloud\n", + " - Also around forty points\n", + " - Could be bucket 0 or 3\n" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "### Integrate Embeddings into Omics Data with SubjectRepresentation\n", + "\n", + "- Let use these omics to enrich the representation of the original dataset.\n", + "\n", + "- The `SubjectRepresentation` function takes our previously generated embeddings and our original Omics Dataset and associated Phenotype\n", + "\n", + "- This function will use the embeddings to enrich the orignal dataset. For more details and how this is performed please view our `GNN Embeddings for Multi-Omics` tab." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "from bioneuralnet.downstream_task import SubjectRepresentation\n", + "\n", + "graph_embed = SubjectRepresentation(\n", + " omics_data=merged_omics,\n", + " embeddings=embeddings,\n", + " phenotype_data=phenotype,\n", + " phenotype_col=\"phenotype\",\n", + " tune=True,\n", + ")\n", + "enhanced_omics = graph_embed.run()" + ] + }, + { + "cell_type": "code", + "execution_count": 20, + "metadata": {}, + "outputs": [ + { + "data": { + "text/plain": [ + "'Before graph embedding:'" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
Gene_1Gene_2Gene_3Gene_4Gene_5Gene_6Gene_7Gene_8Gene_9Gene_10...Mir_91Mir_92Mir_93Mir_94Mir_95Mir_96Mir_97Mir_98Mir_99Mir_100
Samp_122.48570140.35372031.02574520.84720626.69729330.20544923.51200533.67762219.43033330.260153...7.5514625.40787813.93304510.28752110.21331610.60997314.55499610.95565011.42253110.862970
Samp_237.05885034.05223333.48702023.53146126.75462831.73594522.79595229.30153614.93639730.823015...6.8624137.30922613.58618011.97554411.49693710.65374214.41422511.81100412.41366710.719110
Samp_320.53076731.66962335.18956720.95254425.01882632.15723525.06946422.85371918.22022523.092805...6.8749587.75473313.84702912.42440310.93017710.25548413.57035211.31192511.07291511.418794
Samp_433.18688838.48088018.89709731.82330034.04938338.79988724.10646812.39717513.72425527.703085...9.6156236.69359313.84072812.24546610.83689411.50223215.48339910.81281110.12195711.039089
Samp_528.96198141.06049428.49495618.37449530.81523824.00453529.61629624.36404511.40933833.599828...9.1468658.10420613.90345211.4233509.99710710.74458614.46558311.73016612.20615110.724849
\n", + "

5 rows × 600 columns

\n", + "
" + ], + "text/plain": [ + " Gene_1 Gene_2 Gene_3 Gene_4 Gene_5 Gene_6 \\\n", + "Samp_1 22.485701 40.353720 31.025745 20.847206 26.697293 30.205449 \n", + "Samp_2 37.058850 34.052233 33.487020 23.531461 26.754628 31.735945 \n", + "Samp_3 20.530767 31.669623 35.189567 20.952544 25.018826 32.157235 \n", + "Samp_4 33.186888 38.480880 18.897097 31.823300 34.049383 38.799887 \n", + "Samp_5 28.961981 41.060494 28.494956 18.374495 30.815238 24.004535 \n", + "\n", + " Gene_7 Gene_8 Gene_9 Gene_10 ... Mir_91 Mir_92 \\\n", + "Samp_1 23.512005 33.677622 19.430333 30.260153 ... 7.551462 5.407878 \n", + "Samp_2 22.795952 29.301536 14.936397 30.823015 ... 6.862413 7.309226 \n", + "Samp_3 25.069464 22.853719 18.220225 23.092805 ... 6.874958 7.754733 \n", + "Samp_4 24.106468 12.397175 13.724255 27.703085 ... 9.615623 6.693593 \n", + "Samp_5 29.616296 24.364045 11.409338 33.599828 ... 9.146865 8.104206 \n", + "\n", + " Mir_93 Mir_94 Mir_95 Mir_96 Mir_97 Mir_98 \\\n", + "Samp_1 13.933045 10.287521 10.213316 10.609973 14.554996 10.955650 \n", + "Samp_2 13.586180 11.975544 11.496937 10.653742 14.414225 11.811004 \n", + "Samp_3 13.847029 12.424403 10.930177 10.255484 13.570352 11.311925 \n", + "Samp_4 13.840728 12.245466 10.836894 11.502232 15.483399 10.812811 \n", + "Samp_5 13.903452 11.423350 9.997107 10.744586 14.465583 11.730166 \n", + "\n", + " Mir_99 Mir_100 \n", + "Samp_1 11.422531 10.862970 \n", + "Samp_2 12.413667 10.719110 \n", + "Samp_3 11.072915 11.418794 \n", + "Samp_4 10.121957 11.039089 \n", + "Samp_5 12.206151 10.724849 \n", + "\n", + "[5 rows x 600 columns]" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/plain": [ + "'After graph embedding:'" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
Gene_1Gene_2Gene_3Gene_4Gene_5Gene_6Gene_7Gene_8Gene_9Gene_10...Mir_91Mir_92Mir_93Mir_94Mir_95Mir_96Mir_97Mir_98Mir_99Mir_100
Samp_147.351359110.807678-20.252198-11.352853-21.51828652.715821-32.940358-79.9717024.30776081.454875...-17.795733-12.50041416.52891817.119535-8.968622-26.43813930.388110-21.4723430.987792-8.266013
Samp_278.04012393.504362-21.858806-12.814629-21.56449855.386906-31.937167-69.5801423.31144282.969999...-16.171923-16.89541716.11742819.928586-10.095808-26.54720330.094207-23.1487791.073502-8.156545
Samp_343.23457386.961929-22.970151-11.410217-20.16542456.122159-35.122360-54.2689994.03947662.161666...-16.201486-17.92521516.42687620.675535-9.598119-25.55481628.332357-22.1706180.957558-8.688959
Samp_469.886378105.665028-12.335167-17.330151-27.44414467.715195-33.773202-29.4386353.04270674.571709...-22.660122-15.47236816.41940120.377763-9.516204-28.66148832.326442-21.1923880.875321-8.400028
Samp_560.989387112.748415-18.600213-10.006277-24.83739041.893725-41.492480-57.8554562.52948390.444680...-21.555451-18.73302716.49381219.009675-8.778762-26.77357230.201432-22.9903411.055557-8.160912
\n", + "

5 rows × 600 columns

\n", + "
" + ], + "text/plain": [ + " Gene_1 Gene_2 Gene_3 Gene_4 Gene_5 Gene_6 \\\n", + "Samp_1 47.351359 110.807678 -20.252198 -11.352853 -21.518286 52.715821 \n", + "Samp_2 78.040123 93.504362 -21.858806 -12.814629 -21.564498 55.386906 \n", + "Samp_3 43.234573 86.961929 -22.970151 -11.410217 -20.165424 56.122159 \n", + "Samp_4 69.886378 105.665028 -12.335167 -17.330151 -27.444144 67.715195 \n", + "Samp_5 60.989387 112.748415 -18.600213 -10.006277 -24.837390 41.893725 \n", + "\n", + " Gene_7 Gene_8 Gene_9 Gene_10 ... Mir_91 Mir_92 \\\n", + "Samp_1 -32.940358 -79.971702 4.307760 81.454875 ... -17.795733 -12.500414 \n", + "Samp_2 -31.937167 -69.580142 3.311442 82.969999 ... -16.171923 -16.895417 \n", + "Samp_3 -35.122360 -54.268999 4.039476 62.161666 ... -16.201486 -17.925215 \n", + "Samp_4 -33.773202 -29.438635 3.042706 74.571709 ... -22.660122 -15.472368 \n", + "Samp_5 -41.492480 -57.855456 2.529483 90.444680 ... -21.555451 -18.733027 \n", + "\n", + " Mir_93 Mir_94 Mir_95 Mir_96 Mir_97 Mir_98 \\\n", + "Samp_1 16.528918 17.119535 -8.968622 -26.438139 30.388110 -21.472343 \n", + "Samp_2 16.117428 19.928586 -10.095808 -26.547203 30.094207 -23.148779 \n", + "Samp_3 16.426876 20.675535 -9.598119 -25.554816 28.332357 -22.170618 \n", + "Samp_4 16.419401 20.377763 -9.516204 -28.661488 32.326442 -21.192388 \n", + "Samp_5 16.493812 19.009675 -8.778762 -26.773572 30.201432 -22.990341 \n", + "\n", + " Mir_99 Mir_100 \n", + "Samp_1 0.987792 -8.266013 \n", + "Samp_2 1.073502 -8.156545 \n", + "Samp_3 0.957558 -8.688959 \n", + "Samp_4 0.875321 -8.400028 \n", + "Samp_5 1.055557 -8.160912 \n", + "\n", + "[5 rows x 600 columns]" + ] + }, + "metadata": {}, + "output_type": "display_data" + } + ], + "source": [ + "display(\"Before graph embedding:\")\n", + "display(merged_omics.head())\n", + "display(\"After graph embedding:\")\n", + "display(enhanced_omics.head())" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "## Disease Classification with DPMON (Disease Prediction using Multi-Omics Networks)\n", + "\n", + "DPMON extends BioNeuralNet’s embedding pipeline for disease classification tasks. Unlike standard node-level GNN objectives (e.g., regression on network properties), DPMON:\n", + "\n", + "- Combines node embeddings with patient-level omics data, capturing both local and global biological signals.\n", + "- Uses a classification head (e.g., a softmax layer with cross-entropy loss) to predict discrete disease classes.\n", + "- Enables end-to-end training, leveraging GNN-extracted structure and omics context for sample-level prediction.\n", + "\n", + "In this example, we convert the continuous phenotype into 4 discrete classes (0–3) to train a multi-class classifier." + ] + }, + { + "cell_type": "code", + "execution_count": 7, + "metadata": {}, + "outputs": [], + "source": [ + "from bioneuralnet.datasets import DatasetLoader\n", + "import numpy as np\n", + "\n", + "Example = DatasetLoader(\"example1\")\n", + "omics1 = Example.data[\"X1\"]\n", + "omics2 = Example.data[\"X2\"]\n", + "phenotype = Example.data[\"Y\"]\n", + "clinical = Example.data[\"clinical_data\"]\n", + "\n", + "min_val = phenotype[\"phenotype\"].min()\n", + "max_val = phenotype[\"phenotype\"].max()\n", + "\n", + "# linspace creates an array of evenly spaced values\n", + "bins = np.linspace(min_val, max_val, 5)\n", + "\n", + "phenotype[\"phenotype\"] = pd.cut(phenotype[\"phenotype\"], bins=bins, labels=[0, 1, 2, 3], include_lowest=True)\n", + "count_values = phenotype[\"phenotype\"].value_counts(sort=False)" + ] + }, + { + "cell_type": "code", + "execution_count": 8, + "metadata": {}, + "outputs": [ + { + "data": { + "text/html": [ + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
phenotype
Samp_11
Samp_22
Samp_31
Samp_43
Samp_52
......
Samp_3541
Samp_3551
Samp_3562
Samp_3571
Samp_3581
\n", + "

358 rows × 1 columns

\n", + "
" + ], + "text/plain": [ + " phenotype\n", + "Samp_1 1\n", + "Samp_2 2\n", + "Samp_3 1\n", + "Samp_4 3\n", + "Samp_5 2\n", + "... ...\n", + "Samp_354 1\n", + "Samp_355 1\n", + "Samp_356 2\n", + "Samp_357 1\n", + "Samp_358 1\n", + "\n", + "[358 rows x 1 columns]" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/plain": [ + "phenotype\n", + "0 38\n", + "1 158\n", + "2 141\n", + "3 21\n", + "Name: count, dtype: int64" + ] + }, + "metadata": {}, + "output_type": "display_data" + } + ], + "source": [ + "# After binning\n", + "display(phenotype)\n", + "display(count_values)" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "## DPMON Example\n", + "\n", + "In this example, we run **DPMON** in a loop for 3 runs to evaluate its classification performance.\n", + "\n", + "For each run:\n", + "- We instantiate a new `DPMON` model with the same inputs: omics data, phenotype, clinical data, and global network.\n", + "- Set repeat_num = 3, which instructs DPMON to perform three internal iterations. In each iteration it trains an independent model\n", + "- We call `.run()` to generate predictions.\n", + "- We extract the predicted and actual labels.\n", + "- We compute three classification metrics:\n", + " - **Accuracy**\n", + " - **F1-Weighted**\n", + " - **F1-Macro**\n", + "\n", + "After the 3 runs are complete, we calculate the **mean** and **standard deviation** of each metric. These summary statistics allow us to compare DPMON's performance fairly against other models like Random Forest, using consistent metrics and repeated evaluation.\n", + "\n", + "This setup mimics the structure used for other models in the framework, ensuring that DPMON is evaluated in a reproducible and statistically robust way.\n" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "from bioneuralnet.downstream_task import DPMON\n", + "from sklearn.metrics import accuracy_score, f1_score\n", + "import numpy as np\n", + "\n", + "acc_scores = []\n", + "f1w_scores = []\n", + "f1m_scores = []\n", + "\n", + "for i in range(3):\n", + " print(f\"DPMON run {i+1}\")\n", + " \n", + " dpmon = DPMON(\n", + " adjacency_matrix=global_network,\n", + " omics_list=[omics1, omics2],\n", + " phenotype_data=phenotype,\n", + " clinical_data=clinical,\n", + " repeat_num=3,\n", + " tune=True,\n", + " gpu=True,\n", + " cuda=0,\n", + " output_dir=\"dpmon_output\"\n", + " )\n", + " \n", + " predictions_df = dpmon.run()\n", + " y_true = predictions_df[0][\"Actual\"]\n", + " y_pred = predictions_df[0][\"Predicted\"]\n", + "\n", + " acc = accuracy_score(y_true, y_pred)\n", + " f1w = f1_score(y_true, y_pred, average=\"weighted\")\n", + " f1m = f1_score(y_true, y_pred, average=\"macro\")\n", + "\n", + " acc_scores.append(acc)\n", + " f1w_scores.append(f1w)\n", + " f1m_scores.append(f1m)\n", + "\n", + "# get the mean and std in tuple form\n", + "dpmon_acc_tuple = (np.mean(acc_scores), np.std(acc_scores))\n", + "dpmon_f1w_tuple = (np.mean(f1w_scores), np.std(f1w_scores))\n", + "dpmon_f1m_tuple = (np.mean(f1m_scores), np.std(f1m_scores))" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "## Visualizing Results\n", + "\n", + "We now visualize the performance of DPMON compared to a baseline Random Forest using raw omics data.\n", + "\n", + "The metrics shown include:\n", + "- **Accuracy**\n", + "- **F1-Weighted**\n", + "- **F1-Macro**\n", + "\n", + "Each bar represents the mean score across runs, with error bars indicating the standard deviation.\n", + "\n", + "Below, we construct a metrics dictionary and call `plot_multiple_metrics()` to generate the comparison plots.\n" + ] + }, + { + "cell_type": "code", + "execution_count": 10, + "metadata": {}, + "outputs": [ + { + "name": "stderr", + "output_type": "stream", + "text": [ + "2025-05-23 10:47:32,429 - bioneuralnet.metrics.plot - INFO - Plotting multiple metrics: ['Accuracy', 'F1-Weighted', 'F1-Macro']\n" + ] + }, + { + "data": { + "image/png": "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", + "text/plain": [ + "
" + ] + }, + "metadata": {}, + "output_type": "display_data" + } + ], + "source": [ + "from bioneuralnet.metrics import evaluate_rf, plot_multiple_metrics\n", + "\n", + "# raw omics evaluation\n", + "X_raw = merged_omics.values\n", + "y_global = phenotype.values\n", + "rf_acc, rd_f1w, rf_f1m = evaluate_rf(X_raw, y_global, n_estimators=100, runs=5, mode=\"classification\")\n", + "\n", + "# metrics dictionary\n", + "metrics = {\n", + " \"Accuracy\": {\"Raw\": rf_acc,\"DPMON\": dpmon_acc_tuple},\n", + " \"F1-Weighted\": {\"Raw\": rd_f1w,\"DPMON\": dpmon_f1w_tuple},\n", + " \"F1-Macro\": {\"Raw\": rf_f1m, \"DPMON\": dpmon_f1m_tuple}\n", + "}\n", + "\n", + "plot_multiple_metrics(metrics)" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "### Clustering with CorrelatedLouvain and HybridLouvain\n", + "\n", + "- BioNeuralNet includes internal modules for performing correlated clustering on complex networks. \n", + "\n", + "- These methods modify and extend the traditional community detection by integrating phenotype correlation, allowing users to extract biologically relevant, phenotype-associated modules from any network. \n", + "\n", + "- For more details on how this performed, please visit our `Correlated Clustering Methods` tab" + ] + }, + { + "cell_type": "code", + "execution_count": 11, + "metadata": {}, + "outputs": [ + { + "name": "stderr", + "output_type": "stream", + "text": [ + "2025-05-23 10:47:32,549 - bioneuralnet.clustering.hybrid_louvain - INFO - Initializing HybridLouvain...\n", + "2025-05-23 10:47:32,550 - bioneuralnet.clustering.hybrid_louvain - INFO - Initialized HybridLouvain with 600 graph nodes, 600 omics columns, 358 phenotype rows; max_iter=3, k3=0.2, k4=0.8, tune=True\n", + "2025-05-23 10:47:32,551 - bioneuralnet.clustering.hybrid_louvain - INFO - \n", + "Iteration 1/3: Running Correlated Louvain...\n", + "2025-05-23 10:47:32,551 - bioneuralnet.clustering.hybrid_louvain - INFO - Tuning Correlated Louvain for current iteration...\n", + "2025-05-23 10:47:32,561 - bioneuralnet.clustering.correlated_louvain - INFO - Initialized CorrelatedLouvain with k3 = 0.2, k4 = 0.8, \n", + "2025-05-23 10:47:32,561 - bioneuralnet.clustering.correlated_louvain - INFO - Original omics data shape: (358, 600)\n", + "2025-05-23 10:47:32,561 - bioneuralnet.clustering.correlated_louvain - INFO - Original graph has 600 nodes.\n", + "2025-05-23 10:47:32,561 - bioneuralnet.clustering.correlated_louvain - INFO - Final omics data shape: (358, 600)\n", + "2025-05-23 10:47:32,562 - bioneuralnet.clustering.correlated_louvain - INFO - Graph has 600 nodes and 12457 edges.\n", + "2025-05-23 10:47:32,562 - bioneuralnet.clustering.correlated_louvain - INFO - Initialized Correlated Louvain. device=cpu\n", + "2025-05-23 10:47:32,562 - bioneuralnet.clustering.correlated_louvain - INFO - Starting hyperparameter tuning...\n", + "2025-05-23 10:47:32,562\tINFO tune.py:616 -- [output] This uses the legacy output and progress reporter, as Jupyter notebooks are not supported by the new engine, yet. For more information, please see https://github.com/ray-project/ray/issues/36949\n", + "2025-05-23 10:47:37,583\tINFO tune.py:1009 -- Wrote the latest version of all result files and experiment state to '/home/vicente/cl/l' in 0.0113s.\n", + "2025-05-23 10:47:37,597 - bioneuralnet.clustering.correlated_louvain - INFO - Best hyperparameters found: {'k4': 0.9}\n", + "2025-05-23 10:47:37,598 - bioneuralnet.clustering.hybrid_louvain - INFO - Using tuned Louvain parameters: k3=0.09999999999999998, k4=0.9\n", + "2025-05-23 10:47:37,609 - bioneuralnet.clustering.correlated_louvain - INFO - Initialized CorrelatedLouvain with k3 = 0.09999999999999998, k4 = 0.9, \n", + "2025-05-23 10:47:37,609 - bioneuralnet.clustering.correlated_louvain - INFO - Original omics data shape: (358, 600)\n", + "2025-05-23 10:47:37,610 - bioneuralnet.clustering.correlated_louvain - INFO - Original graph has 600 nodes.\n", + "2025-05-23 10:47:37,610 - bioneuralnet.clustering.correlated_louvain - INFO - Final omics data shape: (358, 600)\n", + "2025-05-23 10:47:37,610 - bioneuralnet.clustering.correlated_louvain - INFO - Graph has 600 nodes and 12457 edges.\n", + "2025-05-23 10:47:37,610 - bioneuralnet.clustering.correlated_louvain - INFO - Initialized Correlated Louvain. device=cpu\n", + "2025-05-23 10:47:37,610 - bioneuralnet.clustering.correlated_louvain - INFO - Running standard community detection...\n", + "2025-05-23 10:47:37,666 - bioneuralnet.clustering.correlated_louvain - INFO - Computing community correlation for 83 nodes...\n", + "2025-05-23 10:47:37,667 - bioneuralnet.clustering.correlated_louvain - INFO - B_sub shape: (358, 83), first few columns: ['Gene_13', 'Gene_30', 'Gene_31', 'Gene_42', 'Gene_43']\n", + "2025-05-23 10:47:37,760 - bioneuralnet.clustering.correlated_louvain - INFO - Computing community correlation for 79 nodes...\n", + "2025-05-23 10:47:37,764 - bioneuralnet.clustering.correlated_louvain - INFO - B_sub shape: (358, 79), first few columns: ['Gene_1', 'Gene_2', 'Gene_5', 'Gene_6', 'Gene_7']\n", + "2025-05-23 10:47:37,964 - bioneuralnet.clustering.correlated_louvain - INFO - Computed quality: Q = 0.0019, avg_corr = 0.4264, combined = 0.3840\n", + "2025-05-23 10:47:37,968 - bioneuralnet.clustering.correlated_louvain - INFO - Final quality: 0.3840\n", + "2025-05-23 10:47:37,983 - bioneuralnet.clustering.correlated_louvain - INFO - Computing community correlation for 83 nodes...\n", + "2025-05-23 10:47:37,985 - bioneuralnet.clustering.correlated_louvain - INFO - B_sub shape: (358, 83), first few columns: ['Gene_13', 'Gene_30', 'Gene_31', 'Gene_42', 'Gene_43']\n", + "2025-05-23 10:47:38,102 - bioneuralnet.clustering.correlated_louvain - INFO - Computing community correlation for 79 nodes...\n", + "2025-05-23 10:47:38,104 - bioneuralnet.clustering.correlated_louvain - INFO - B_sub shape: (358, 79), first few columns: ['Gene_1', 'Gene_2', 'Gene_5', 'Gene_6', 'Gene_7']\n", + "2025-05-23 10:47:38,140 - bioneuralnet.clustering.correlated_louvain - INFO - Computed quality: Q = 0.0019, avg_corr = 0.4264, combined = 0.3840\n", + "2025-05-23 10:47:38,140 - bioneuralnet.clustering.hybrid_louvain - INFO - Iteration 1: Louvain Quality = 0.3840\n", + "2025-05-23 10:47:38,141 - bioneuralnet.clustering.correlated_louvain - INFO - Computing community correlation for 83 nodes...\n", + "2025-05-23 10:47:38,142 - bioneuralnet.clustering.correlated_louvain - INFO - B_sub shape: (358, 83), first few columns: ['Gene_13', 'Gene_30', 'Gene_31', 'Gene_42', 'Gene_43']\n", + "2025-05-23 10:47:38,154 - bioneuralnet.clustering.correlated_louvain - INFO - Computing community correlation for 79 nodes...\n", + "2025-05-23 10:47:38,155 - bioneuralnet.clustering.correlated_louvain - INFO - B_sub shape: (358, 79), first few columns: ['Gene_1', 'Gene_2', 'Gene_5', 'Gene_6', 'Gene_7']\n", + "2025-05-23 10:47:38,162 - bioneuralnet.clustering.hybrid_louvain - INFO - Selected seed community of size 79 with correlation -0.7265\n", + "2025-05-23 10:47:38,162 - bioneuralnet.clustering.hybrid_louvain - INFO - Tuning Correlated PageRank for current iteration...\n", + "2025-05-23 10:47:38,163 - bioneuralnet.clustering.correlated_pagerank - INFO - Initialized PageRank with the following parameters:\n", + "2025-05-23 10:47:38,163 - bioneuralnet.clustering.correlated_pagerank - INFO - Graph: NetworkX Graph with 600 nodes and 12457 edges.\n", + "2025-05-23 10:47:38,164 - bioneuralnet.clustering.correlated_pagerank - INFO - Omics Data: DataFrame with shape (358, 600).\n", + "2025-05-23 10:47:38,164 - bioneuralnet.clustering.correlated_pagerank - INFO - Phenotype Data: Series with 358 samples.\n", + "2025-05-23 10:47:38,164 - bioneuralnet.clustering.correlated_pagerank - INFO - Alpha: 0.9\n", + "2025-05-23 10:47:38,164 - bioneuralnet.clustering.correlated_pagerank - INFO - Max Iterations: 100\n", + "2025-05-23 10:47:38,165 - bioneuralnet.clustering.correlated_pagerank - INFO - Tolerance: 1e-06\n", + "2025-05-23 10:47:38,165 - bioneuralnet.clustering.correlated_pagerank - INFO - K (Composite Score Weight): 0.5\n", + "2025-05-23 10:47:38,166 - bioneuralnet.clustering.correlated_pagerank - INFO - Initialized Correlated Louvain. device=cpu\n", + "2025-05-23 10:47:38,166\tINFO tune.py:616 -- [output] This uses the legacy output and progress reporter, as Jupyter notebooks are not supported by the new engine, yet. For more information, please see https://github.com/ray-project/ray/issues/36949\n", + "2025-05-23 10:48:46,763\tINFO tune.py:1009 -- Wrote the latest version of all result files and experiment state to '/home/vicente/pr/l' in 0.0186s.\n", + "2025-05-23 10:48:46,809 - bioneuralnet.clustering.correlated_pagerank - INFO - Best hyperparameters found: {'alpha': 0.8, 'k': 0.5, 'max_iter': 500, 'tol': 0.00013762486088297676}\n", + "2025-05-23 10:48:46,811 - bioneuralnet.clustering.hybrid_louvain - INFO - Using tuned PageRank parameters: alpha=0.8, max_iter=500, tol=0.00013762486088297676, k=0.5\n", + "2025-05-23 10:48:46,811 - bioneuralnet.clustering.correlated_pagerank - INFO - Initialized PageRank with the following parameters:\n", + "2025-05-23 10:48:46,812 - bioneuralnet.clustering.correlated_pagerank - INFO - Graph: NetworkX Graph with 600 nodes and 12457 edges.\n", + "2025-05-23 10:48:46,812 - bioneuralnet.clustering.correlated_pagerank - INFO - Omics Data: DataFrame with shape (358, 600).\n", + "2025-05-23 10:48:46,812 - bioneuralnet.clustering.correlated_pagerank - INFO - Phenotype Data: Series with 358 samples.\n", + "2025-05-23 10:48:46,812 - bioneuralnet.clustering.correlated_pagerank - INFO - Alpha: 0.8\n", + "2025-05-23 10:48:46,812 - bioneuralnet.clustering.correlated_pagerank - INFO - Max Iterations: 500\n", + "2025-05-23 10:48:46,812 - bioneuralnet.clustering.correlated_pagerank - INFO - Tolerance: 0.00013762486088297676\n", + "2025-05-23 10:48:46,812 - bioneuralnet.clustering.correlated_pagerank - INFO - K (Composite Score Weight): 0.5\n", + "2025-05-23 10:48:46,813 - bioneuralnet.clustering.correlated_pagerank - INFO - Initialized Correlated Louvain. device=cpu\n", + "2025-05-23 10:48:47,325 - bioneuralnet.clustering.correlated_pagerank - INFO - Generated personalization vector for seed nodes: ['Gene_1', 'Gene_2', 'Gene_5', 'Gene_6', 'Gene_7', 'Gene_9', 'Gene_10', 'Gene_21', 'Gene_23', 'Gene_24', 'Gene_26', 'Gene_28', 'Gene_45', 'Gene_46', 'Gene_53', 'Gene_54', 'Gene_56', 'Gene_57', 'Gene_72', 'Gene_74', 'Gene_77', 'Gene_78', 'Gene_79', 'Gene_82', 'Gene_83', 'Gene_86', 'Gene_88', 'Gene_90', 'Gene_123', 'Gene_136', 'Gene_142', 'Gene_168', 'Gene_170', 'Gene_174', 'Gene_183', 'Gene_189', 'Gene_190', 'Gene_194', 'Gene_200', 'Gene_206', 'Gene_214', 'Gene_216', 'Gene_217', 'Gene_219', 'Gene_222', 'Gene_228', 'Gene_229', 'Gene_237', 'Gene_244', 'Gene_250', 'Gene_268', 'Gene_287', 'Gene_288', 'Gene_299', 'Gene_308', 'Gene_324', 'Gene_335', 'Gene_352', 'Gene_367', 'Gene_370', 'Gene_372', 'Gene_385', 'Gene_391', 'Gene_406', 'Gene_407', 'Gene_411', 'Gene_416', 'Gene_445', 'Gene_446', 'Gene_447', 'Gene_457', 'Gene_474', 'Gene_478', 'Gene_481', 'Gene_485', 'Gene_490', 'Gene_500', 'Mir_2', 'Mir_96']\n", + "2025-05-23 10:48:47,335 - bioneuralnet.clustering.correlated_pagerank - INFO - PageRank computation completed.\n", + "2025-05-23 10:48:48,959 - bioneuralnet.clustering.correlated_pagerank - WARNING - Skipping iteration 161 due to zero volume (vol_S=46.027221978868376, vol_T=0).\n", + "2025-05-23 10:48:48,966 - bioneuralnet.clustering.correlated_pagerank - INFO - Sweep cut resulted in cluster of size 35 with conductance 0.56 and correlation 0.73.\n", + "2025-05-23 10:48:48,966 - bioneuralnet.clustering.correlated_pagerank - INFO - PageRank clustering completed successfully.\n", + "2025-05-23 10:48:48,967 - bioneuralnet.clustering.hybrid_louvain - INFO - Refined subgraph size: 35\n", + "2025-05-23 10:48:48,969 - bioneuralnet.clustering.hybrid_louvain - INFO - \n", + "Iteration 2/3: Running Correlated Louvain...\n", + "2025-05-23 10:48:48,969 - bioneuralnet.clustering.hybrid_louvain - INFO - Tuning Correlated Louvain for current iteration...\n", + "2025-05-23 10:48:48,971 - bioneuralnet.clustering.correlated_louvain - INFO - Initialized CorrelatedLouvain with k3 = 0.2, k4 = 0.8, \n", + "2025-05-23 10:48:48,972 - bioneuralnet.clustering.correlated_louvain - INFO - Original omics data shape: (358, 600)\n", + "2025-05-23 10:48:48,972 - bioneuralnet.clustering.correlated_louvain - INFO - Original graph has 35 nodes.\n", + "2025-05-23 10:48:48,972 - bioneuralnet.clustering.correlated_louvain - INFO - Final omics data shape: (358, 600)\n", + "2025-05-23 10:48:48,972 - bioneuralnet.clustering.correlated_louvain - INFO - Graph has 35 nodes and 594 edges.\n", + "2025-05-23 10:48:48,972 - bioneuralnet.clustering.correlated_louvain - INFO - Initialized Correlated Louvain. device=cpu\n", + "2025-05-23 10:48:48,973 - bioneuralnet.clustering.correlated_louvain - INFO - Starting hyperparameter tuning...\n", + "2025-05-23 10:48:48,973\tINFO tune.py:616 -- [output] This uses the legacy output and progress reporter, as Jupyter notebooks are not supported by the new engine, yet. For more information, please see https://github.com/ray-project/ray/issues/36949\n", + "2025-05-23 10:48:53,902\tINFO tune.py:1009 -- Wrote the latest version of all result files and experiment state to '/home/vicente/cl/l' in 0.0108s.\n", + "2025-05-23 10:48:53,916 - bioneuralnet.clustering.correlated_louvain - INFO - Best hyperparameters found: {'k4': 0.9}\n", + "2025-05-23 10:48:53,917 - bioneuralnet.clustering.hybrid_louvain - INFO - Using tuned Louvain parameters: k3=0.09999999999999998, k4=0.9\n", + "2025-05-23 10:48:53,918 - bioneuralnet.clustering.correlated_louvain - INFO - Initialized CorrelatedLouvain with k3 = 0.09999999999999998, k4 = 0.9, \n", + "2025-05-23 10:48:53,918 - bioneuralnet.clustering.correlated_louvain - INFO - Original omics data shape: (358, 600)\n", + "2025-05-23 10:48:53,919 - bioneuralnet.clustering.correlated_louvain - INFO - Original graph has 35 nodes.\n", + "2025-05-23 10:48:53,919 - bioneuralnet.clustering.correlated_louvain - INFO - Final omics data shape: (358, 600)\n", + "2025-05-23 10:48:53,919 - bioneuralnet.clustering.correlated_louvain - INFO - Graph has 35 nodes and 594 edges.\n", + "2025-05-23 10:48:53,919 - bioneuralnet.clustering.correlated_louvain - INFO - Initialized Correlated Louvain. device=cpu\n", + "2025-05-23 10:48:53,920 - bioneuralnet.clustering.correlated_louvain - INFO - Running standard community detection...\n", + "2025-05-23 10:48:53,922 - bioneuralnet.clustering.correlated_louvain - INFO - Computing community correlation for 31 nodes...\n", + "2025-05-23 10:48:53,924 - bioneuralnet.clustering.correlated_louvain - INFO - B_sub shape: (358, 31), first few columns: ['Gene_445', 'Gene_335', 'Gene_53', 'Gene_190', 'Gene_299']\n", + "2025-05-23 10:48:53,926 - bioneuralnet.clustering.correlated_louvain - INFO - Computing community correlation for 4 nodes...\n", + "2025-05-23 10:48:53,926 - bioneuralnet.clustering.correlated_louvain - INFO - B_sub shape: (358, 4), first few columns: ['Gene_255', 'Gene_313', 'Gene_114', 'Gene_281']\n", + "2025-05-23 10:48:53,928 - bioneuralnet.clustering.correlated_louvain - INFO - Computed quality: Q = 0.0000, avg_corr = 0.4263, combined = 0.3837\n", + "2025-05-23 10:48:53,929 - bioneuralnet.clustering.correlated_louvain - INFO - Final quality: 0.3837\n", + "2025-05-23 10:48:53,929 - bioneuralnet.clustering.correlated_louvain - INFO - Computing community correlation for 31 nodes...\n", + "2025-05-23 10:48:53,930 - bioneuralnet.clustering.correlated_louvain - INFO - B_sub shape: (358, 31), first few columns: ['Gene_445', 'Gene_335', 'Gene_53', 'Gene_190', 'Gene_299']\n", + "2025-05-23 10:48:53,932 - bioneuralnet.clustering.correlated_louvain - INFO - Computing community correlation for 4 nodes...\n", + "2025-05-23 10:48:53,932 - bioneuralnet.clustering.correlated_louvain - INFO - B_sub shape: (358, 4), first few columns: ['Gene_255', 'Gene_313', 'Gene_114', 'Gene_281']\n", + "2025-05-23 10:48:53,934 - bioneuralnet.clustering.correlated_louvain - INFO - Computed quality: Q = 0.0000, avg_corr = 0.4263, combined = 0.3837\n", + "2025-05-23 10:48:53,934 - bioneuralnet.clustering.hybrid_louvain - INFO - Iteration 2: Louvain Quality = 0.3837\n", + "2025-05-23 10:48:53,934 - bioneuralnet.clustering.correlated_louvain - INFO - Computing community correlation for 31 nodes...\n", + "2025-05-23 10:48:53,935 - bioneuralnet.clustering.correlated_louvain - INFO - B_sub shape: (358, 31), first few columns: ['Gene_445', 'Gene_335', 'Gene_53', 'Gene_190', 'Gene_299']\n", + "2025-05-23 10:48:53,937 - bioneuralnet.clustering.correlated_louvain - INFO - Computing community correlation for 4 nodes...\n", + "2025-05-23 10:48:53,938 - bioneuralnet.clustering.correlated_louvain - INFO - B_sub shape: (358, 4), first few columns: ['Gene_255', 'Gene_313', 'Gene_114', 'Gene_281']\n", + "2025-05-23 10:48:53,939 - bioneuralnet.clustering.hybrid_louvain - INFO - Selected seed community of size 31 with correlation 0.7215\n", + "2025-05-23 10:48:53,940 - bioneuralnet.clustering.hybrid_louvain - INFO - Tuning Correlated PageRank for current iteration...\n", + "2025-05-23 10:48:53,940 - bioneuralnet.clustering.correlated_pagerank - INFO - Initialized PageRank with the following parameters:\n", + "2025-05-23 10:48:53,940 - bioneuralnet.clustering.correlated_pagerank - INFO - Graph: NetworkX Graph with 35 nodes and 594 edges.\n", + "2025-05-23 10:48:53,940 - bioneuralnet.clustering.correlated_pagerank - INFO - Omics Data: DataFrame with shape (358, 600).\n", + "2025-05-23 10:48:53,941 - bioneuralnet.clustering.correlated_pagerank - INFO - Phenotype Data: Series with 358 samples.\n", + "2025-05-23 10:48:53,941 - bioneuralnet.clustering.correlated_pagerank - INFO - Alpha: 0.9\n", + "2025-05-23 10:48:53,941 - bioneuralnet.clustering.correlated_pagerank - INFO - Max Iterations: 100\n", + "2025-05-23 10:48:53,941 - bioneuralnet.clustering.correlated_pagerank - INFO - Tolerance: 1e-06\n", + "2025-05-23 10:48:53,941 - bioneuralnet.clustering.correlated_pagerank - INFO - K (Composite Score Weight): 0.5\n", + "2025-05-23 10:48:53,941 - bioneuralnet.clustering.correlated_pagerank - INFO - Initialized Correlated Louvain. device=cpu\n", + "2025-05-23 10:48:53,942\tINFO tune.py:616 -- [output] This uses the legacy output and progress reporter, as Jupyter notebooks are not supported by the new engine, yet. For more information, please see https://github.com/ray-project/ray/issues/36949\n", + "2025-05-23 10:49:11,398\tINFO tune.py:1009 -- Wrote the latest version of all result files and experiment state to '/home/vicente/pr/l' in 0.0273s.\n", + "2025-05-23 10:49:11,438 - bioneuralnet.clustering.correlated_pagerank - INFO - Best hyperparameters found: {'alpha': 0.8, 'k': 0.5, 'max_iter': 500, 'tol': 4.2498925749875235e-05}\n", + "2025-05-23 10:49:11,440 - bioneuralnet.clustering.hybrid_louvain - INFO - Using tuned PageRank parameters: alpha=0.8, max_iter=500, tol=4.2498925749875235e-05, k=0.5\n", + "2025-05-23 10:49:11,441 - bioneuralnet.clustering.correlated_pagerank - INFO - Initialized PageRank with the following parameters:\n", + "2025-05-23 10:49:11,441 - bioneuralnet.clustering.correlated_pagerank - INFO - Graph: NetworkX Graph with 35 nodes and 594 edges.\n", + "2025-05-23 10:49:11,441 - bioneuralnet.clustering.correlated_pagerank - INFO - Omics Data: DataFrame with shape (358, 600).\n", + "2025-05-23 10:49:11,441 - bioneuralnet.clustering.correlated_pagerank - INFO - Phenotype Data: Series with 358 samples.\n", + "2025-05-23 10:49:11,441 - bioneuralnet.clustering.correlated_pagerank - INFO - Alpha: 0.8\n", + "2025-05-23 10:49:11,441 - bioneuralnet.clustering.correlated_pagerank - INFO - Max Iterations: 500\n", + "2025-05-23 10:49:11,441 - bioneuralnet.clustering.correlated_pagerank - INFO - Tolerance: 4.2498925749875235e-05\n", + "2025-05-23 10:49:11,441 - bioneuralnet.clustering.correlated_pagerank - INFO - K (Composite Score Weight): 0.5\n", + "2025-05-23 10:49:11,442 - bioneuralnet.clustering.correlated_pagerank - INFO - Initialized Correlated Louvain. device=cpu\n", + "2025-05-23 10:49:11,500 - bioneuralnet.clustering.correlated_pagerank - INFO - Generated personalization vector for seed nodes: ['Gene_445', 'Gene_335', 'Gene_53', 'Gene_190', 'Gene_299', 'Gene_142', 'Gene_23', 'Gene_45', 'Mir_96', 'Gene_77', 'Gene_474', 'Gene_500', 'Gene_28', 'Gene_407', 'Gene_2', 'Gene_6', 'Gene_288', 'Gene_174', 'Gene_1', 'Gene_7', 'Gene_183', 'Gene_485', 'Gene_446', 'Gene_411', 'Gene_88', 'Gene_79', 'Gene_54', 'Gene_385', 'Gene_367', 'Gene_46', 'Gene_447']\n", + "2025-05-23 10:49:11,501 - bioneuralnet.clustering.correlated_pagerank - INFO - PageRank computation completed.\n", + "2025-05-23 10:49:11,561 - bioneuralnet.clustering.correlated_pagerank - INFO - Sweep cut resulted in cluster of size 5 with conductance 0.569 and correlation -0.69.\n", + "2025-05-23 10:49:11,561 - bioneuralnet.clustering.correlated_pagerank - INFO - PageRank clustering completed successfully.\n", + "2025-05-23 10:49:11,561 - bioneuralnet.clustering.hybrid_louvain - INFO - Refined subgraph size: 5\n", + "2025-05-23 10:49:11,562 - bioneuralnet.clustering.hybrid_louvain - INFO - \n", + "Iteration 3/3: Running Correlated Louvain...\n", + "2025-05-23 10:49:11,562 - bioneuralnet.clustering.hybrid_louvain - INFO - Tuning Correlated Louvain for current iteration...\n", + "2025-05-23 10:49:11,564 - bioneuralnet.clustering.correlated_louvain - INFO - Initialized CorrelatedLouvain with k3 = 0.2, k4 = 0.8, \n", + "2025-05-23 10:49:11,564 - bioneuralnet.clustering.correlated_louvain - INFO - Original omics data shape: (358, 600)\n", + "2025-05-23 10:49:11,565 - bioneuralnet.clustering.correlated_louvain - INFO - Original graph has 5 nodes.\n", + "2025-05-23 10:49:11,565 - bioneuralnet.clustering.correlated_louvain - INFO - Final omics data shape: (358, 600)\n", + "2025-05-23 10:49:11,565 - bioneuralnet.clustering.correlated_louvain - INFO - Graph has 5 nodes and 10 edges.\n", + "2025-05-23 10:49:11,565 - bioneuralnet.clustering.correlated_louvain - INFO - Initialized Correlated Louvain. device=cpu\n", + "2025-05-23 10:49:11,565 - bioneuralnet.clustering.correlated_louvain - INFO - Starting hyperparameter tuning...\n", + "2025-05-23 10:49:11,565\tINFO tune.py:616 -- [output] This uses the legacy output and progress reporter, as Jupyter notebooks are not supported by the new engine, yet. For more information, please see https://github.com/ray-project/ray/issues/36949\n", + "2025-05-23 10:49:16,307\tINFO tune.py:1009 -- Wrote the latest version of all result files and experiment state to '/home/vicente/cl/l' in 0.0102s.\n", + "2025-05-23 10:49:16,322 - bioneuralnet.clustering.correlated_louvain - INFO - Best hyperparameters found: {'k4': 0.9}\n", + "2025-05-23 10:49:16,322 - bioneuralnet.clustering.hybrid_louvain - INFO - Using tuned Louvain parameters: k3=0.09999999999999998, k4=0.9\n", + "2025-05-23 10:49:16,324 - bioneuralnet.clustering.correlated_louvain - INFO - Initialized CorrelatedLouvain with k3 = 0.09999999999999998, k4 = 0.9, \n", + "2025-05-23 10:49:16,324 - bioneuralnet.clustering.correlated_louvain - INFO - Original omics data shape: (358, 600)\n", + "2025-05-23 10:49:16,324 - bioneuralnet.clustering.correlated_louvain - INFO - Original graph has 5 nodes.\n", + "2025-05-23 10:49:16,324 - bioneuralnet.clustering.correlated_louvain - INFO - Final omics data shape: (358, 600)\n", + "2025-05-23 10:49:16,325 - bioneuralnet.clustering.correlated_louvain - INFO - Graph has 5 nodes and 10 edges.\n", + "2025-05-23 10:49:16,325 - bioneuralnet.clustering.correlated_louvain - INFO - Initialized Correlated Louvain. device=cpu\n", + "2025-05-23 10:49:16,325 - bioneuralnet.clustering.correlated_louvain - INFO - Running standard community detection...\n", + "2025-05-23 10:49:16,325 - bioneuralnet.clustering.correlated_louvain - INFO - Computing community correlation for 5 nodes...\n", + "2025-05-23 10:49:16,327 - bioneuralnet.clustering.correlated_louvain - INFO - B_sub shape: (358, 5), first few columns: ['Gene_7', 'Gene_53', 'Gene_446', 'Gene_6', 'Gene_1']\n", + "2025-05-23 10:49:16,329 - bioneuralnet.clustering.correlated_louvain - INFO - Computed quality: Q = 0.0000, avg_corr = 0.6920, combined = 0.6228\n", + "2025-05-23 10:49:16,330 - bioneuralnet.clustering.correlated_louvain - INFO - Final quality: 0.6228\n", + "2025-05-23 10:49:16,330 - bioneuralnet.clustering.correlated_louvain - INFO - Computing community correlation for 5 nodes...\n", + "2025-05-23 10:49:16,331 - bioneuralnet.clustering.correlated_louvain - INFO - B_sub shape: (358, 5), first few columns: ['Gene_7', 'Gene_53', 'Gene_446', 'Gene_6', 'Gene_1']\n", + "2025-05-23 10:49:16,332 - bioneuralnet.clustering.correlated_louvain - INFO - Computed quality: Q = 0.0000, avg_corr = 0.6920, combined = 0.6228\n", + "2025-05-23 10:49:16,332 - bioneuralnet.clustering.hybrid_louvain - INFO - Iteration 3: Louvain Quality = 0.6228\n", + "2025-05-23 10:49:16,333 - bioneuralnet.clustering.correlated_louvain - INFO - Computing community correlation for 5 nodes...\n", + "2025-05-23 10:49:16,333 - bioneuralnet.clustering.correlated_louvain - INFO - B_sub shape: (358, 5), first few columns: ['Gene_7', 'Gene_53', 'Gene_446', 'Gene_6', 'Gene_1']\n", + "2025-05-23 10:49:16,335 - bioneuralnet.clustering.hybrid_louvain - INFO - Selected seed community of size 5 with correlation -0.6920\n", + "2025-05-23 10:49:16,335 - bioneuralnet.clustering.hybrid_louvain - INFO - Tuning Correlated PageRank for current iteration...\n", + "2025-05-23 10:49:16,335 - bioneuralnet.clustering.correlated_pagerank - INFO - Initialized PageRank with the following parameters:\n", + "2025-05-23 10:49:16,335 - bioneuralnet.clustering.correlated_pagerank - INFO - Graph: NetworkX Graph with 5 nodes and 10 edges.\n", + "2025-05-23 10:49:16,336 - bioneuralnet.clustering.correlated_pagerank - INFO - Omics Data: DataFrame with shape (358, 600).\n", + "2025-05-23 10:49:16,336 - bioneuralnet.clustering.correlated_pagerank - INFO - Phenotype Data: Series with 358 samples.\n", + "2025-05-23 10:49:16,336 - bioneuralnet.clustering.correlated_pagerank - INFO - Alpha: 0.9\n", + "2025-05-23 10:49:16,336 - bioneuralnet.clustering.correlated_pagerank - INFO - Max Iterations: 100\n", + "2025-05-23 10:49:16,336 - bioneuralnet.clustering.correlated_pagerank - INFO - Tolerance: 1e-06\n", + "2025-05-23 10:49:16,336 - bioneuralnet.clustering.correlated_pagerank - INFO - K (Composite Score Weight): 0.5\n", + "2025-05-23 10:49:16,336 - bioneuralnet.clustering.correlated_pagerank - INFO - Initialized Correlated Louvain. device=cpu\n", + "2025-05-23 10:49:16,336\tINFO tune.py:616 -- [output] This uses the legacy output and progress reporter, as Jupyter notebooks are not supported by the new engine, yet. For more information, please see https://github.com/ray-project/ray/issues/36949\n", + "2025-05-23 10:49:32,784\tINFO tune.py:1009 -- Wrote the latest version of all result files and experiment state to '/home/vicente/pr/l' in 0.0266s.\n", + "2025-05-23 10:49:32,823 - bioneuralnet.clustering.correlated_pagerank - INFO - Best hyperparameters found: {'alpha': 0.8, 'k': 0.5, 'max_iter': 1000, 'tol': 6.934060257491069e-05}\n", + "2025-05-23 10:49:32,825 - bioneuralnet.clustering.hybrid_louvain - INFO - Using tuned PageRank parameters: alpha=0.8, max_iter=1000, tol=6.934060257491069e-05, k=0.5\n", + "2025-05-23 10:49:32,826 - bioneuralnet.clustering.correlated_pagerank - INFO - Initialized PageRank with the following parameters:\n", + "2025-05-23 10:49:32,826 - bioneuralnet.clustering.correlated_pagerank - INFO - Graph: NetworkX Graph with 5 nodes and 10 edges.\n", + "2025-05-23 10:49:32,826 - bioneuralnet.clustering.correlated_pagerank - INFO - Omics Data: DataFrame with shape (358, 600).\n", + "2025-05-23 10:49:32,826 - bioneuralnet.clustering.correlated_pagerank - INFO - Phenotype Data: Series with 358 samples.\n", + "2025-05-23 10:49:32,827 - bioneuralnet.clustering.correlated_pagerank - INFO - Alpha: 0.8\n", + "2025-05-23 10:49:32,827 - bioneuralnet.clustering.correlated_pagerank - INFO - Max Iterations: 1000\n", + "2025-05-23 10:49:32,827 - bioneuralnet.clustering.correlated_pagerank - INFO - Tolerance: 6.934060257491069e-05\n", + "2025-05-23 10:49:32,827 - bioneuralnet.clustering.correlated_pagerank - INFO - K (Composite Score Weight): 0.5\n", + "2025-05-23 10:49:32,827 - bioneuralnet.clustering.correlated_pagerank - INFO - Initialized Correlated Louvain. device=cpu\n", + "2025-05-23 10:49:32,837 - bioneuralnet.clustering.correlated_pagerank - INFO - Generated personalization vector for seed nodes: ['Gene_7', 'Gene_53', 'Gene_446', 'Gene_6', 'Gene_1']\n", + "2025-05-23 10:49:32,837 - bioneuralnet.clustering.correlated_pagerank - INFO - PageRank computation completed.\n", + "2025-05-23 10:49:32,844 - bioneuralnet.clustering.correlated_pagerank - WARNING - No valid sweep cut found. Returning empty cluster.\n", + "2025-05-23 10:49:32,844 - bioneuralnet.clustering.correlated_pagerank - WARNING - Sweep cut did not identify any cluster.\n", + "2025-05-23 10:49:32,844 - bioneuralnet.clustering.correlated_pagerank - INFO - PageRank clustering completed successfully.\n", + "2025-05-23 10:49:32,844 - bioneuralnet.clustering.hybrid_louvain - INFO - Refined subgraph size: 0\n", + "2025-05-23 10:49:32,844 - bioneuralnet.clustering.hybrid_louvain - INFO - Subgraph size converged or too small. Stopping iterations.\n", + "2025-05-23 10:49:32,844 - bioneuralnet.clustering.hybrid_louvain - INFO - Hybrid Louvain completed after 3 iterations.\n" + ] + }, + { + "data": { + "text/plain": [ + "'Number of clusters:'" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/plain": [ + "2" + ] + }, + "metadata": {}, + "output_type": "display_data" + } + ], + "source": [ + "from bioneuralnet.clustering import HybridLouvain\n", + "import networkx as nx\n", + "\n", + "merged_omics = pd.concat([omics1, omics2], axis=1)\n", + "G_network = nx.from_pandas_adjacency(global_network)\n", + "\n", + "hybrid = HybridLouvain(\n", + " G=G_network,\n", + " B=merged_omics,\n", + " Y=phenotype,\n", + " tune=True,\n", + ")\n", + "hybrid_result = hybrid.run(as_dfs=True)\n", + "display(\"Number of clusters:\", len(hybrid_result))" + ] + }, + { + "cell_type": "code", + "execution_count": 13, + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + " phenotype\n", + "Samp_1 1\n", + "Samp_2 2\n", + "Samp_3 1\n", + "Samp_4 3\n", + "Samp_5 2\n", + "... ...\n", + "Samp_354 1\n", + "Samp_355 1\n", + "Samp_356 2\n", + "Samp_357 1\n", + "Samp_358 1\n", + "\n", + "[358 rows x 1 columns]\n" + ] + }, + { + "data": { + "image/png": "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", + "text/plain": [ + "
" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
ClusterLouvain SizeLouvain CorrelationSMCCNET SizeSMCCNET Correlation
0Cluster_1350.821929650.641804
1Cluster_250.800173400.067307
\n", + "
" + ], + "text/plain": [ + " Cluster Louvain Size Louvain Correlation SMCCNET Size \\\n", + "0 Cluster_1 35 0.821929 65 \n", + "1 Cluster_2 5 0.800173 40 \n", + "\n", + " SMCCNET Correlation \n", + "0 0.641804 \n", + "1 0.067307 " + ] + }, + "execution_count": 13, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "# Lets compare hytbrid louvain with the SmCCNet clusters\n", + "import matplotlib.pyplot as plt\n", + "\n", + "print(phenotype)\n", + "\n", + "compare_clusters(hybrid_result, clusters, phenotype, merged_omics)" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "### Lets plot the clustered network from correlated louvain" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [ + { + "data": { + "image/png": "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", + "text/plain": [ + "
" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
OmicDegree
Index
25Gene_4116
21Gene_75
19Gene_1745
20Gene_14
8Gene_1424
\n", + "
" + ], + "text/plain": [ + " Omic Degree\n", + "Index \n", + "25 Gene_411 6\n", + "21 Gene_7 5\n", + "19 Gene_174 5\n", + "20 Gene_1 4\n", + "8 Gene_142 4" + ] + }, + "metadata": {}, + "output_type": "display_data" + } + ], + "source": [ + "from bioneuralnet.metrics import plot_network\n", + "from bioneuralnet.metrics import louvain_to_adjacency\n", + "\n", + "cluster1 = hybrid_result[0]\n", + "cluster2 = hybrid_result[1]\n", + "\n", + "# Convert Louvain clusters into adjacency matrices\n", + "louvain_adj1 = louvain_to_adjacency(cluster1)\n", + "louvain_adj2 = louvain_to_adjacency(cluster2)\n", + "\n", + "# Plot using the converted adjacency matrices\n", + "\n", + "cluster1_mapping = plot_network(louvain_adj1, weight_threshold=0.1, show_labels=True, show_edge_weights=False)\n", + "display(cluster1_mapping.head())\n" + ] + }, + { + "cell_type": "code", + "execution_count": 19, + "metadata": {}, + "outputs": [ + { + "data": { + "image/png": "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", + "text/plain": [ + "
" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
OmicDegree
Index
1Gene_74
2Gene_4464
3Gene_64
4Gene_534
5Gene_14
\n", + "
" + ], + "text/plain": [ + " Omic Degree\n", + "Index \n", + "1 Gene_7 4\n", + "2 Gene_446 4\n", + "3 Gene_6 4\n", + "4 Gene_53 4\n", + "5 Gene_1 4" + ] + }, + "metadata": {}, + "output_type": "display_data" + } + ], + "source": [ + "cluster2_mapping = plot_network(louvain_adj2, weight_threshold=0.01, show_labels=True, show_edge_weights=True)\n", + "display(cluster2_mapping.head())" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "#### Package Version" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "BioNeuralNet version: 1.0.7\n" + ] + } + ], + "source": [ + "import bioneuralnet\n", + "print(\"BioNeuralNet version:\", bioneuralnet.__version__)" + ] + } + ], + "metadata": { + "kernelspec": { + "display_name": ".enviroment", + "language": "python", + "name": "python3" + }, + "language_info": { + "codemirror_mode": { + "name": "ipython", + "version": 3 + }, + "file_extension": ".py", + "mimetype": "text/x-python", + "name": "python", + "nbconvert_exporter": "python", + "pygments_lexer": "ipython3", + "version": "3.12.3" + } + }, + "nbformat": 4, + "nbformat_minor": 2 +} diff --git a/docs/source/Quick_Start.ipynb b/docs/source/Quick_Start.ipynb index 0bc7175..533cdee 100644 --- a/docs/source/Quick_Start.ipynb +++ b/docs/source/Quick_Start.ipynb @@ -8,59 +8,6 @@ "This notebook demonstrates the core functionality of the BioNeuralNet package. It covers data loading, network generation, network embedding via GNNs, subject representation, downstream disease prediction, evaluation metrics, clustering, and use of external tools.\"" ] }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "from bioneuralnet.downstream_task import SubjectRepresentation\n", - "from bioneuralnet.downstream_task import DPMON\n", - "from bioneuralnet.clustering import CorrelatedPageRank\n", - "from bioneuralnet.clustering import CorrelatedLouvain\n", - "from bioneuralnet.clustering import HybridLouvain\n", - "\n", - "from bioneuralnet.utils import get_logger\n", - "from bioneuralnet.utils import rdata_to_df\n", - "from bioneuralnet.utils import variance_summary\n", - "from bioneuralnet.utils import zero_fraction_summary\n", - "from bioneuralnet.utils import expression_summary\n", - "from bioneuralnet.utils import correlation_summary\n", - "from bioneuralnet.utils import explore_data_stats\n", - "from bioneuralnet.utils import preprocess_clinical\n", - "from bioneuralnet.utils import clean_inf_nan\n", - "from bioneuralnet.utils import select_top_k_variance\n", - "from bioneuralnet.utils import select_top_k_correlation\n", - "from bioneuralnet.utils import select_top_randomforest\n", - "from bioneuralnet.utils import top_anova_f_features\n", - "from bioneuralnet.utils import prune_network\n", - "from bioneuralnet.utils import prune_network_by_quantile\n", - "from bioneuralnet.utils import network_remove_low_variance\n", - "from bioneuralnet.utils import network_remove_high_zero_fraction\n", - "from bioneuralnet.utils import gen_similarity_graph\n", - "from bioneuralnet.utils import gen_correlation_graph\n", - "from bioneuralnet.utils import gen_threshold_graph\n", - "from bioneuralnet.utils import gen_gaussian_knn_graph\n", - "from bioneuralnet.utils import gen_lasso_graph\n", - "from bioneuralnet.utils import gen_mst_graph\n", - "from bioneuralnet.utils import gen_snn_graph\n", - "\n", - "from bioneuralnet.metrics import omics_correlation\n", - "from bioneuralnet.metrics import cluster_correlation\n", - "from bioneuralnet.metrics import louvain_to_adjacency\n", - "from bioneuralnet.metrics import evaluate_rf\n", - "from bioneuralnet.metrics import plot_performance_three\n", - "from bioneuralnet.metrics import plot_variance_distribution\n", - "from bioneuralnet.metrics import plot_variance_by_feature\n", - "from bioneuralnet.metrics import plot_performance\n", - "from bioneuralnet.metrics import plot_embeddings\n", - "from bioneuralnet.metrics import plot_network\n", - "from bioneuralnet.metrics import compare_clusters\n", - "\n", - "from bioneuralnet.datasets import DatasetLoader\n", - "from bioneuralnet.external_tools import SmCCNet\n" - ] - }, { "cell_type": "markdown", "metadata": {}, diff --git a/docs/source/datasets.ipynb b/docs/source/datasets.ipynb new file mode 100644 index 0000000..0ae5cf1 --- /dev/null +++ b/docs/source/datasets.ipynb @@ -0,0 +1,2260 @@ +{ + "cells": [ + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "# Datasets Guide\n", + "\n", + "\n", + "- The `DatasetLoader` class provides a simple interface to access example multi-omics datasets included in this package. \n", + "- Each dataset is loaded as a collection of **pandas DataFrames**, with table names as keys and the corresponding data as values. \n", + "- Users can explore the structure of any dataset via the `.shape` property, which returns a mapping from table name to `(rows, columns)`. \n", + "- Three datasets are available out-of-the-box:\n", + "\n", + " 1. **example1**:\n", + " - Synthetic dataset designed for testing and demonstration \n", + " - Contains small DataFrames: `X1`, `X2`, `Y`, `clinical_data` \n", + " - Useful for quick checks of package functionality\n", + "\n", + " 2. **monet**: \n", + " - Multi-omics benchmark dataset from the **Multi-Omics NETwork Analysis Workshop (MONET)**. \n", + " - Includes multiple DataFrames: `gene_data`, `mirna_data`, `phenotype`, `rppa_data`, `clinical_data` \n", + " - Workshop details: \n", + "\n", + " 3. **brca** :\n", + " - Breast cancer cohort from TCGA (BRCA project) \n", + " - Provides comprehensive omics DataFrames: `rna`, `mirna`, `meth`, `pam50`, `clinical` \n", + " - Full dataset description available at: \n", + "\n" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "### Shapes overview:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [ + { + "data": { + "text/plain": [ + "'example1 shapes'" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/plain": [ + "' - X1: 358 x 500'" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/plain": [ + "' - X2: 358 x 100'" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/plain": [ + "' - Y: 358 x 1'" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/plain": [ + "' - clinical_data: 358 x 6'" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/plain": [ + "'monet shapes'" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/plain": [ + "' - gene_data: 107 x 5039'" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/plain": [ + "' - mirna_data: 107 x 789'" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/plain": [ + "' - phenotype: 106 x 1'" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/plain": [ + "' - rppa_data: 107 x 175'" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/plain": [ + "' - clinical_data: 107 x 5'" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/plain": [ + "'brca shapes'" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/plain": [ + "' - mirna: 769 x 503'" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/plain": [ + "' - pam50: 769 x 1'" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/plain": [ + "' - clinical: 769 x 118'" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/plain": [ + "' - rna: 769 x 6000'" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/plain": [ + "' - meth: 769 x 6000'" + ] + }, + "metadata": {}, + "output_type": "display_data" + } + ], + "source": [ + "from bioneuralnet.datasets import DatasetLoader\n", + "\n", + "for name in [\"example1\", \"monet\", \"brca\"]:\n", + " ds = DatasetLoader(name)\n", + " display(f\"{name} shapes:\")\n", + " for tbl, (rows, cols) in ds.shape.items():\n", + " display(f\" - {tbl}: {rows} x {cols}\")" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "### Example 1: Synthetic dataset" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [ + { + "data": { + "text/html": [ + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
Gene_1Gene_2Gene_3Gene_4Gene_5
Samp_122.48570140.35372031.02574520.84720626.697293
Samp_237.05885034.05223333.48702023.53146126.754628
Samp_320.53076731.66962335.18956720.95254425.018826
Samp_433.18688838.48088018.89709731.82330034.049383
Samp_528.96198141.06049428.49495618.37449530.815238
..................
Samp_35424.52065228.59540931.29966632.09537933.659730
Samp_35531.25278928.98808729.57419531.18928832.098841
Samp_35624.89482625.94488730.85264126.70515830.102546
Samp_35717.03433738.57470525.09520137.06244235.417758
Samp_35820.83916727.09978831.03845319.41085931.818995
\n", + "

358 rows × 5 columns

\n", + "
" + ], + "text/plain": [ + " Gene_1 Gene_2 Gene_3 Gene_4 Gene_5\n", + "Samp_1 22.485701 40.353720 31.025745 20.847206 26.697293\n", + "Samp_2 37.058850 34.052233 33.487020 23.531461 26.754628\n", + "Samp_3 20.530767 31.669623 35.189567 20.952544 25.018826\n", + "Samp_4 33.186888 38.480880 18.897097 31.823300 34.049383\n", + "Samp_5 28.961981 41.060494 28.494956 18.374495 30.815238\n", + "... ... ... ... ... ...\n", + "Samp_354 24.520652 28.595409 31.299666 32.095379 33.659730\n", + "Samp_355 31.252789 28.988087 29.574195 31.189288 32.098841\n", + "Samp_356 24.894826 25.944887 30.852641 26.705158 30.102546\n", + "Samp_357 17.034337 38.574705 25.095201 37.062442 35.417758\n", + "Samp_358 20.839167 27.099788 31.038453 19.410859 31.818995\n", + "\n", + "[358 rows x 5 columns]" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
Mir_1Mir_2Mir_3Mir_4Mir_5
Samp_115.22391317.54582615.78471914.89198310.348205
Samp_216.30696516.67283013.36152914.48854912.660905
Samp_316.54511916.73500514.61747217.84526713.822790
Samp_413.98689916.20743216.29307817.72528612.300565
Samp_516.33833217.39386916.39792515.85372513.387675
..................
Samp_35415.06506516.07983014.63561617.01384511.612843
Samp_35515.99757615.44895115.35556616.50175211.701778
Samp_35615.20686214.39537816.21800116.04495513.650741
Samp_35714.47412915.48286315.51254915.13661314.531277
Samp_35815.09418816.04730415.29887117.02266515.046043
\n", + "

358 rows × 5 columns

\n", + "
" + ], + "text/plain": [ + " Mir_1 Mir_2 Mir_3 Mir_4 Mir_5\n", + "Samp_1 15.223913 17.545826 15.784719 14.891983 10.348205\n", + "Samp_2 16.306965 16.672830 13.361529 14.488549 12.660905\n", + "Samp_3 16.545119 16.735005 14.617472 17.845267 13.822790\n", + "Samp_4 13.986899 16.207432 16.293078 17.725286 12.300565\n", + "Samp_5 16.338332 17.393869 16.397925 15.853725 13.387675\n", + "... ... ... ... ... ...\n", + "Samp_354 15.065065 16.079830 14.635616 17.013845 11.612843\n", + "Samp_355 15.997576 15.448951 15.355566 16.501752 11.701778\n", + "Samp_356 15.206862 14.395378 16.218001 16.044955 13.650741\n", + "Samp_357 14.474129 15.482863 15.512549 15.136613 14.531277\n", + "Samp_358 15.094188 16.047304 15.298871 17.022665 15.046043\n", + "\n", + "[358 rows x 5 columns]" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
phenotype
Samp_1235.067423
Samp_2253.544991
Samp_3234.204994
Samp_4281.035429
Samp_5245.447781
......
Samp_354236.120451
Samp_355222.572359
Samp_356268.472285
Samp_357235.808167
Samp_358213.886123
\n", + "

358 rows × 1 columns

\n", + "
" + ], + "text/plain": [ + " phenotype\n", + "Samp_1 235.067423\n", + "Samp_2 253.544991\n", + "Samp_3 234.204994\n", + "Samp_4 281.035429\n", + "Samp_5 245.447781\n", + "... ...\n", + "Samp_354 236.120451\n", + "Samp_355 222.572359\n", + "Samp_356 268.472285\n", + "Samp_357 235.808167\n", + "Samp_358 213.886123\n", + "\n", + "[358 rows x 1 columns]" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
AgeGenderBMIChronic_BronchitisEmphysema
PatientID
Samp_178031.211
Samp_268119.210
Samp_354119.301
Samp_447136.200
Samp_560126.201
..................
Samp_35471023.010
Samp_35562125.501
Samp_35661021.110
Samp_35764037.600
Samp_35861131.310
\n", + "

358 rows × 5 columns

\n", + "
" + ], + "text/plain": [ + " Age Gender BMI Chronic_Bronchitis Emphysema\n", + "PatientID \n", + "Samp_1 78 0 31.2 1 1\n", + "Samp_2 68 1 19.2 1 0\n", + "Samp_3 54 1 19.3 0 1\n", + "Samp_4 47 1 36.2 0 0\n", + "Samp_5 60 1 26.2 0 1\n", + "... ... ... ... ... ...\n", + "Samp_354 71 0 23.0 1 0\n", + "Samp_355 62 1 25.5 0 1\n", + "Samp_356 61 0 21.1 1 0\n", + "Samp_357 64 0 37.6 0 0\n", + "Samp_358 61 1 31.3 1 0\n", + "\n", + "[358 rows x 5 columns]" + ] + }, + "metadata": {}, + "output_type": "display_data" + } + ], + "source": [ + "from bioneuralnet.datasets import DatasetLoader\n", + "\n", + "Example = DatasetLoader(\"example1\")\n", + "omics1 = Example.data[\"X1\"]\n", + "omics2 = Example.data[\"X2\"]\n", + "phenotype = Example.data[\"Y\"]\n", + "clinical = Example.data[\"clinical_data\"]\n", + "\n", + "display(omics1.iloc[:, :5])\n", + "display(omics2.iloc[:, :5])\n", + "display(phenotype.iloc[:, :5])\n", + "display(clinical.iloc[:, :5])" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "### Monet: Set from the **Multi-Omics NETwork Analysis Workshop (MONET)**, Univ. of Colorado Anschutz " + ] + }, + { + "cell_type": "code", + "execution_count": 5, + "metadata": {}, + "outputs": [ + { + "data": { + "text/html": [ + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
A2ML1AACSLAADACAADATAATK
00.4666710.0748450.990309-0.4108731.897562
1-0.524465-0.146727-0.735206-0.6284560.170962
2-0.029879-0.626509-0.735206-0.6778920.020060
30.674895-0.626509-0.330409-0.662162-0.911966
4-0.110607-0.626509-0.735206-0.8485420.042645
..................
1020.9996001.343979-0.7352061.7426740.253103
103-0.919337-0.626509-0.735206-1.380461-0.899019
104-0.606702-0.6265090.497658-0.717505-0.625122
1051.9113460.021380-0.1666240.7855420.344953
106-0.444363-0.111055-0.735206-1.395805-0.685468
\n", + "

107 rows × 5 columns

\n", + "
" + ], + "text/plain": [ + " A2ML1 AACSL AADAC AADAT AATK\n", + "0 0.466671 0.074845 0.990309 -0.410873 1.897562\n", + "1 -0.524465 -0.146727 -0.735206 -0.628456 0.170962\n", + "2 -0.029879 -0.626509 -0.735206 -0.677892 0.020060\n", + "3 0.674895 -0.626509 -0.330409 -0.662162 -0.911966\n", + "4 -0.110607 -0.626509 -0.735206 -0.848542 0.042645\n", + ".. ... ... ... ... ...\n", + "102 0.999600 1.343979 -0.735206 1.742674 0.253103\n", + "103 -0.919337 -0.626509 -0.735206 -1.380461 -0.899019\n", + "104 -0.606702 -0.626509 0.497658 -0.717505 -0.625122\n", + "105 1.911346 0.021380 -0.166624 0.785542 0.344953\n", + "106 -0.444363 -0.111055 -0.735206 -1.395805 -0.685468\n", + "\n", + "[107 rows x 5 columns]" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
hsa-let-7a-1hsa-let-7a-2hsa-let-7a-3hsa-let-7bhsa-let-7c
0-0.832527-0.851616-0.837155-1.079659-0.181270
10.2291550.2496960.2342950.859289-0.057729
20.4142680.4170230.4089130.6350591.195203
3-0.855214-0.869152-0.862713-1.955447-0.572552
41.3653101.3562521.3517501.259095-0.316760
..................
102-1.402001-1.401125-1.349961-1.5343861.231456
1032.5512772.5470462.5631911.0547690.981436
1040.1821380.1887300.191094-1.0606150.345907
1050.2894890.2924700.297778-0.1978501.040270
1060.7910680.8006800.8036580.5895980.937994
\n", + "

107 rows × 5 columns

\n", + "
" + ], + "text/plain": [ + " hsa-let-7a-1 hsa-let-7a-2 hsa-let-7a-3 hsa-let-7b hsa-let-7c\n", + "0 -0.832527 -0.851616 -0.837155 -1.079659 -0.181270\n", + "1 0.229155 0.249696 0.234295 0.859289 -0.057729\n", + "2 0.414268 0.417023 0.408913 0.635059 1.195203\n", + "3 -0.855214 -0.869152 -0.862713 -1.955447 -0.572552\n", + "4 1.365310 1.356252 1.351750 1.259095 -0.316760\n", + ".. ... ... ... ... ...\n", + "102 -1.402001 -1.401125 -1.349961 -1.534386 1.231456\n", + "103 2.551277 2.547046 2.563191 1.054769 0.981436\n", + "104 0.182138 0.188730 0.191094 -1.060615 0.345907\n", + "105 0.289489 0.292470 0.297778 -0.197850 1.040270\n", + "106 0.791068 0.800680 0.803658 0.589598 0.937994\n", + "\n", + "[107 rows x 5 columns]" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
0
01
10
20
30
40
......
1010
1021
1030
1040
1050
\n", + "

106 rows × 1 columns

\n", + "
" + ], + "text/plain": [ + " 0\n", + "0 1\n", + "1 0\n", + "2 0\n", + "3 0\n", + "4 0\n", + ".. ..\n", + "101 0\n", + "102 1\n", + "103 0\n", + "104 0\n", + "105 0\n", + "\n", + "[106 rows x 1 columns]" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
YWHAEYWHAZEIF4EBP1TP53BP1ARAF
0-0.3579980.099812-1.067285-0.412211-0.357998
1-0.055031-0.5174450.032633-0.743096-0.055031
2-0.137863-0.5596900.302764-0.968388-0.137863
3-0.170726-0.028206-0.3414610.282581-0.170726
4-1.430765-0.138087-0.545894-0.616864-1.430765
..................
102-0.708685-0.7788131.623365-0.090612-0.708685
1030.261442-0.407563-0.567735-0.1869190.261442
1041.3508661.461061-1.159541-1.6748741.350866
1050.179510-0.300029-1.048938-0.6216800.179510
106-1.099437-0.818285-0.855339-1.313032-1.099437
\n", + "

107 rows × 5 columns

\n", + "
" + ], + "text/plain": [ + " YWHAE YWHAZ EIF4EBP1 TP53BP1 ARAF\n", + "0 -0.357998 0.099812 -1.067285 -0.412211 -0.357998\n", + "1 -0.055031 -0.517445 0.032633 -0.743096 -0.055031\n", + "2 -0.137863 -0.559690 0.302764 -0.968388 -0.137863\n", + "3 -0.170726 -0.028206 -0.341461 0.282581 -0.170726\n", + "4 -1.430765 -0.138087 -0.545894 -0.616864 -1.430765\n", + ".. ... ... ... ... ...\n", + "102 -0.708685 -0.778813 1.623365 -0.090612 -0.708685\n", + "103 0.261442 -0.407563 -0.567735 -0.186919 0.261442\n", + "104 1.350866 1.461061 -1.159541 -1.674874 1.350866\n", + "105 0.179510 -0.300029 -1.048938 -0.621680 0.179510\n", + "106 -1.099437 -0.818285 -0.855339 -1.313032 -1.099437\n", + "\n", + "[107 rows x 5 columns]" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
overall_survivalstatusyears_to_birthraceradiation_therapy
03015037blackorafricanamericanyes
12348173whiteyes
23011041asianyes
33283067whiteno
41873042whiteno
..................
1022329063whiteyes
1031004174whiteyes
104984046whiteyes
105867044whiteno
1061133053whiteyes
\n", + "

107 rows × 5 columns

\n", + "
" + ], + "text/plain": [ + " overall_survival status years_to_birth race \\\n", + "0 3015 0 37 blackorafricanamerican \n", + "1 2348 1 73 white \n", + "2 3011 0 41 asian \n", + "3 3283 0 67 white \n", + "4 1873 0 42 white \n", + ".. ... ... ... ... \n", + "102 2329 0 63 white \n", + "103 1004 1 74 white \n", + "104 984 0 46 white \n", + "105 867 0 44 white \n", + "106 1133 0 53 white \n", + "\n", + " radiation_therapy \n", + "0 yes \n", + "1 yes \n", + "2 yes \n", + "3 no \n", + "4 no \n", + ".. ... \n", + "102 yes \n", + "103 yes \n", + "104 yes \n", + "105 no \n", + "106 yes \n", + "\n", + "[107 rows x 5 columns]" + ] + }, + "metadata": {}, + "output_type": "display_data" + } + ], + "source": [ + "from bioneuralnet.datasets import DatasetLoader\n", + "\n", + "monet = DatasetLoader(\"monet\")\n", + "gene = monet.data[\"gene_data\"]\n", + "mirna = monet.data[\"mirna_data\"]\n", + "phenotype = monet.data[\"phenotype\"]\n", + "rppa = monet.data[\"rppa_data\"]\n", + "clinical = monet.data[\"clinical_data\"]\n", + "\n", + "display(gene.iloc[:, :5])\n", + "display(mirna.iloc[:, :5])\n", + "display(phenotype.iloc[:, :5])\n", + "display(rppa.iloc[:, :5])\n", + "display(clinical.iloc[:, :5])" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "### BRCA: Breast cancer cohort dataset." + ] + }, + { + "cell_type": "code", + "execution_count": 6, + "metadata": {}, + "outputs": [ + { + "data": { + "text/html": [ + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
ESR1_2099FOXA1_3169MLPH_79083AGR3_155465TBC1D9_23158
patient
TCGA-3C-AAAU11.75570612.41160812.56677512.04972914.173402
TCGA-3C-AALI6.09835812.56259614.10126312.43169111.295692
TCGA-3C-AALJ12.86927012.17371712.31543511.49609812.314665
TCGA-3C-AALK11.27921112.84393913.37929110.15357112.610953
TCGA-4H-AAAK12.43000812.73122912.58092010.25367212.353710
..................
TCGA-WT-AB4412.15442112.58394915.22331211.01516411.632502
TCGA-XX-A89911.41547612.00154713.0672129.70433912.868580
TCGA-XX-A89A11.28757611.98877112.76982510.19002511.919563
TCGA-Z7-A8R511.68885211.54486112.52218610.55692412.289650
TCGA-Z7-A8R612.65058612.97678712.83372411.66858712.534803
\n", + "

769 rows × 5 columns

\n", + "
" + ], + "text/plain": [ + " ESR1_2099 FOXA1_3169 MLPH_79083 AGR3_155465 TBC1D9_23158\n", + "patient \n", + "TCGA-3C-AAAU 11.755706 12.411608 12.566775 12.049729 14.173402\n", + "TCGA-3C-AALI 6.098358 12.562596 14.101263 12.431691 11.295692\n", + "TCGA-3C-AALJ 12.869270 12.173717 12.315435 11.496098 12.314665\n", + "TCGA-3C-AALK 11.279211 12.843939 13.379291 10.153571 12.610953\n", + "TCGA-4H-AAAK 12.430008 12.731229 12.580920 10.253672 12.353710\n", + "... ... ... ... ... ...\n", + "TCGA-WT-AB44 12.154421 12.583949 15.223312 11.015164 11.632502\n", + "TCGA-XX-A899 11.415476 12.001547 13.067212 9.704339 12.868580\n", + "TCGA-XX-A89A 11.287576 11.988771 12.769825 10.190025 11.919563\n", + "TCGA-Z7-A8R5 11.688852 11.544861 12.522186 10.556924 12.289650\n", + "TCGA-Z7-A8R6 12.650586 12.976787 12.833724 11.668587 12.534803\n", + "\n", + "[769 rows x 5 columns]" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
hsa_let_7a_1hsa_let_7a_2hsa_let_7a_3hsa_let_7bhsa_let_7c
patient
TCGA-3C-AAAU13.12976514.11793313.14771414.5951358.414890
TCGA-3C-AALI12.91806913.92230012.91319414.5126579.646536
TCGA-3C-AALJ13.01203314.01000213.02848313.4196129.312455
TCGA-3C-AALK13.14469714.14172113.15128114.66719611.511431
TCGA-4H-AAAK13.41168414.41351813.42048114.43854811.693927
..................
TCGA-WT-AB4413.37571514.36667113.36982714.51402411.926315
TCGA-XX-A89914.03615515.03634114.04331314.33950312.361761
TCGA-XX-A89A13.67956914.68485513.69146314.19820712.684212
TCGA-Z7-A8R512.96208813.96635012.98489714.32066011.980246
TCGA-Z7-A8R613.34971114.34990113.37844714.11721210.378041
\n", + "

769 rows × 5 columns

\n", + "
" + ], + "text/plain": [ + " hsa_let_7a_1 hsa_let_7a_2 hsa_let_7a_3 hsa_let_7b hsa_let_7c\n", + "patient \n", + "TCGA-3C-AAAU 13.129765 14.117933 13.147714 14.595135 8.414890\n", + "TCGA-3C-AALI 12.918069 13.922300 12.913194 14.512657 9.646536\n", + "TCGA-3C-AALJ 13.012033 14.010002 13.028483 13.419612 9.312455\n", + "TCGA-3C-AALK 13.144697 14.141721 13.151281 14.667196 11.511431\n", + "TCGA-4H-AAAK 13.411684 14.413518 13.420481 14.438548 11.693927\n", + "... ... ... ... ... ...\n", + "TCGA-WT-AB44 13.375715 14.366671 13.369827 14.514024 11.926315\n", + "TCGA-XX-A899 14.036155 15.036341 14.043313 14.339503 12.361761\n", + "TCGA-XX-A89A 13.679569 14.684855 13.691463 14.198207 12.684212\n", + "TCGA-Z7-A8R5 12.962088 13.966350 12.984897 14.320660 11.980246\n", + "TCGA-Z7-A8R6 13.349711 14.349901 13.378447 14.117212 10.378041\n", + "\n", + "[769 rows x 5 columns]" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
SFT2D2IL17RAMIR128_1FOXA1LOC145837
patient
TCGA-3C-AAAU-2.196646-0.0497423.355022-3.934344-1.801595
TCGA-3C-AALI-2.436039-0.2170062.652026-3.995267-1.512691
TCGA-3C-AALJ-2.390041-0.3601802.564778-3.917724-0.701434
TCGA-3C-AALK-2.469813-0.1077912.718057-4.100320-0.756467
TCGA-4H-AAAK-2.501687-0.0917743.086157-3.628072-0.090305
..................
TCGA-WT-AB44-2.358699-0.0928633.138854-3.864208-0.446164
TCGA-XX-A899-2.633115-0.1926983.330302-3.4984190.144114
TCGA-XX-A89A-2.602103-0.2877182.287165-3.720622-0.236061
TCGA-Z7-A8R5-2.572044-0.1467913.000648-3.3356910.693710
TCGA-Z7-A8R6-2.357329-0.1407414.496094-4.857515-1.210885
\n", + "

769 rows × 5 columns

\n", + "
" + ], + "text/plain": [ + " SFT2D2 IL17RA MIR128_1 FOXA1 LOC145837\n", + "patient \n", + "TCGA-3C-AAAU -2.196646 -0.049742 3.355022 -3.934344 -1.801595\n", + "TCGA-3C-AALI -2.436039 -0.217006 2.652026 -3.995267 -1.512691\n", + "TCGA-3C-AALJ -2.390041 -0.360180 2.564778 -3.917724 -0.701434\n", + "TCGA-3C-AALK -2.469813 -0.107791 2.718057 -4.100320 -0.756467\n", + "TCGA-4H-AAAK -2.501687 -0.091774 3.086157 -3.628072 -0.090305\n", + "... ... ... ... ... ...\n", + "TCGA-WT-AB44 -2.358699 -0.092863 3.138854 -3.864208 -0.446164\n", + "TCGA-XX-A899 -2.633115 -0.192698 3.330302 -3.498419 0.144114\n", + "TCGA-XX-A89A -2.602103 -0.287718 2.287165 -3.720622 -0.236061\n", + "TCGA-Z7-A8R5 -2.572044 -0.146791 3.000648 -3.335691 0.693710\n", + "TCGA-Z7-A8R6 -2.357329 -0.140741 4.496094 -4.857515 -1.210885\n", + "\n", + "[769 rows x 5 columns]" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
pam50
patient
TCGA-3C-AAAU3
TCGA-3C-AALI2
TCGA-3C-AALJ4
TCGA-3C-AALK3
TCGA-4H-AAAK3
......
TCGA-WT-AB443
TCGA-XX-A8993
TCGA-XX-A89A3
TCGA-Z7-A8R53
TCGA-Z7-A8R64
\n", + "

769 rows × 1 columns

\n", + "
" + ], + "text/plain": [ + " pam50\n", + "patient \n", + "TCGA-3C-AAAU 3\n", + "TCGA-3C-AALI 2\n", + "TCGA-3C-AALJ 4\n", + "TCGA-3C-AALK 3\n", + "TCGA-4H-AAAK 3\n", + "... ...\n", + "TCGA-WT-AB44 3\n", + "TCGA-XX-A899 3\n", + "TCGA-XX-A89A 3\n", + "TCGA-Z7-A8R5 3\n", + "TCGA-Z7-A8R6 4\n", + "\n", + "[769 rows x 1 columns]" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
synchronous_malignancyajcc_pathologic_stagedays_to_diagnosislateralitycreated_datetime
patient
TCGA-3C-AAAUNoStage X0.0LeftNaN
TCGA-3C-AALINoStage IIB0.0RightNaN
TCGA-3C-AALJNoStage IIB0.0RightNaN
TCGA-3C-AALKNoStage IA0.0RightNaN
TCGA-4H-AAAKNoStage IIIA0.0LeftNaN
..................
TCGA-WT-AB44NoStage IA0.0LeftNaN
TCGA-XX-A899NoStage IIIA0.0RightNaN
TCGA-XX-A89ANoStage IIB0.0LeftNaN
TCGA-Z7-A8R5NoStage IIIA0.0LeftNaN
TCGA-Z7-A8R6NoStage I0.0LeftNaN
\n", + "

769 rows × 5 columns

\n", + "
" + ], + "text/plain": [ + " synchronous_malignancy ajcc_pathologic_stage days_to_diagnosis \\\n", + "patient \n", + "TCGA-3C-AAAU No Stage X 0.0 \n", + "TCGA-3C-AALI No Stage IIB 0.0 \n", + "TCGA-3C-AALJ No Stage IIB 0.0 \n", + "TCGA-3C-AALK No Stage IA 0.0 \n", + "TCGA-4H-AAAK No Stage IIIA 0.0 \n", + "... ... ... ... \n", + "TCGA-WT-AB44 No Stage IA 0.0 \n", + "TCGA-XX-A899 No Stage IIIA 0.0 \n", + "TCGA-XX-A89A No Stage IIB 0.0 \n", + "TCGA-Z7-A8R5 No Stage IIIA 0.0 \n", + "TCGA-Z7-A8R6 No Stage I 0.0 \n", + "\n", + " laterality created_datetime \n", + "patient \n", + "TCGA-3C-AAAU Left NaN \n", + "TCGA-3C-AALI Right NaN \n", + "TCGA-3C-AALJ Right NaN \n", + "TCGA-3C-AALK Right NaN \n", + "TCGA-4H-AAAK Left NaN \n", + "... ... ... \n", + "TCGA-WT-AB44 Left NaN \n", + "TCGA-XX-A899 Right NaN \n", + "TCGA-XX-A89A Left NaN \n", + "TCGA-Z7-A8R5 Left NaN \n", + "TCGA-Z7-A8R6 Left NaN \n", + "\n", + "[769 rows x 5 columns]" + ] + }, + "metadata": {}, + "output_type": "display_data" + } + ], + "source": [ + "from bioneuralnet.datasets import DatasetLoader\n", + "\n", + "brca = DatasetLoader(\"brca\")\n", + "rna = brca.data[\"rna\"]\n", + "mirna = brca.data[\"mirna\"]\n", + "meth = brca.data[\"meth\"]\n", + "pam50 = brca.data[\"pam50\"]\n", + "clinical = brca.data[\"clinical\"]\n", + "\n", + "display(rna.iloc[:, :5])\n", + "display(mirna.iloc[:, :5])\n", + "display(meth.iloc[:, :5])\n", + "display(pam50.iloc[:, :5])\n", + "display(clinical.iloc[:, :5])" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "## Generating an Omics Network\n", + "Try generating a network with one of our datasets (e.g., `\"monet\"`, `\"example1\"`, or `\"brca\"`).\n", + "\n", + "Supported graph construction methods include:\n", + "\n", + "- **Cosine similarity / RBF kernel graphs** (`gen_similarity_graph`) \n", + "- **Pearson / Spearman correlation graphs** (`gen_correlation_graph`) \n", + "- **Soft-threshold (WGCNA-style) graphs** (`gen_threshold_graph`) \n", + "- **Gaussian k-NN graphs** (`gen_gaussian_knn_graph`) \n", + "- **Mutual information graphs** (`gen_mutual_info_graph`) \n", + "- **Graphical Lasso (sparse inverse covariance) graphs** (`gen_lasso_graph`) \n", + "- **Minimum Spanning Tree (MST) graphs** (`gen_mst_graph`) \n", + "- **Shared Nearest Neighbor (SNN) graphs** (`gen_snn_graph`)\n", + "\n", + "For more details on all of these utilities, see the [utils documentation](https://bioneuralnet.readthedocs.io/en/latest/utils.html). " + ] + } + ], + "metadata": { + "kernelspec": { + "display_name": ".enviroment", + "language": "python", + "name": "python3" + }, + "language_info": { + "codemirror_mode": { + "name": "ipython", + "version": 3 + }, + "file_extension": ".py", + "mimetype": "text/x-python", + "name": "python", + "nbconvert_exporter": "python", + "pygments_lexer": "ipython3", + "version": "3.12.3" + } + }, + "nbformat": 4, + "nbformat_minor": 2 +} diff --git a/docs/source/index.rst b/docs/source/index.rst index d85c805..89bc9b6 100644 --- a/docs/source/index.rst +++ b/docs/source/index.rst @@ -200,6 +200,7 @@ We welcome contributions to BioNeuralNet! If you have ideas for new features, im metrics utils downstream_tasks + datasets.ipynb Quick_Start.ipynb TCGA-BRCA_Dataset.ipynb tutorials/index