Skip to content

README.md

Latest commit

 

History

History
35 lines (23 loc) · 1.67 KB

README.md

File metadata and controls

35 lines (23 loc) · 1.67 KB

Leukemia Classification Project

This project is dedicated to the classification of leukemia types based on gene expression profiles. It utilizes a dataset of gene expressions for patients diagnosed with different types of leukemia. The main objective is to build a predictive model that can accurately classify the type of leukemia from the gene expression data, trained on very small data.

Dataset

The dataset used in this project is Leukemia_GSE28497.csv, which contains gene expression data for various leukemia types. The data is hosted at https://sbcb.inf.ufrgs.br/data/cumida/Genes/Leukemia/GSE28497/.

Dataset Structure

The dataset includes the following columns:

  • sample: The sample code of the patient (used as an index for the data).
  • type: The type of leukemia diagnosed.
  • Gene_name1, Gene_name2, ..., Gene_nameN: Expression levels of thousands of genes.

Dataset Class

A Python class named Dataset is used to manage the dataset. This class handles downloading the data (if not present), loading it into a Pandas DataFrame, and performing initial preprocessing including:

  • Encoding the type column using label encoding.
  • Retaining the sample code within the DataFrame without setting it as an index.

Usage outside the scope of the project

To use the Dataset class, ensure you have Python installed along with the Pandas library. Here is a basic example of how to initialize the dataset and access the data:

# Initialize the dataset
dataset = Dataset()

# Access the processed DataFrame
data_frame = dataset.data

# Get the mapping of leukemia types
label_mapping = dataset.get_label_mapping()
print("Leukemia Type Mapping:", label_mapping)