cNF Modularization And Call Scripts #3
Conversation
|
Should be ready - let me know if you have any questions. |
sgosline
left a comment
sgosline
left a comment
There was a problem hiding this comment.
Just a few documentation issues - otherwise it's pretty close.
1- please add comments to helper files and descriptions/context in the R markdowns.
2- clearly separate out scripts/functions from analysis. The scripts/functions should be data/batch agnostic (which I think they are) . The analysis sections can be re-run as we get more analyses, but shoudl also explain what you are doing, why you are doing it, and maybe provide some support for the result.
3- no binary files in github if you can avoid it. this includes docx.
There was a problem hiding this comment.
Github doesn't really support binary files - can you please remove? Maybe you can move to an R MD (maybe create a separate 'analysis' folder and date the analysis you did.
There was a problem hiding this comment.
can you please add some comments to the functions so it's clear what each of them does?
There was a problem hiding this comment.
is this something that would be run with every analysis/batch? or is it a one-off? if the latter - move to an analysis folder, date it, and add some more context/description. If it something to be run with every batch, then move it to a .R file.
There was a problem hiding this comment.
move to an analysis folder, proide more context/description.
…being tested so there may be more updates
…some to legacy. Added comments and cleaned things up. Moved the docx file to analysis readme
| ## Quick run order (and what each file sources) | ||
|
|
||
| 1) **01_run_normalize_omics.Rmd** | ||
| - *Sources:* `cNF_helper_code.R`, `01_normalize_batchcorrect_omics.R` |
There was a problem hiding this comment.
This code actually doesn't run - you should be sourcing files using relative path names, e.g. ../cNF_helper_code.R.
| 1) **01_run_normalize_omics.Rmd** | ||
| - *Sources:* `cNF_helper_code.R`, `01_normalize_batchcorrect_omics.R` | ||
| - Runs per-modality normalization and ComBat; writes long tables/plots if enabled. | ||
|
|
There was a problem hiding this comment.
Why would I run this? How do I enable writing the table? When would I do this?
| 2) **02_analyze_modality.Rmd** | ||
| - *Sources:* `cNF_helper_code.R`, `02_analyze_modality_correlations.R` | ||
| - Builds drug/feature matrices, modality correlations. | ||
|
|
There was a problem hiding this comment.
What is a modality correlation? Why do I need these matrices?
| 3) **03_pathway_enrichment.Rmd** | ||
| - *Sources:* `cNF_helper_code.R`, `03_leapr_biomarker.R` | ||
| - LeapR pathway enrichment and plots for all. | ||
|
|
There was a problem hiding this comment.
I'd add here that the goal is pathway enrichment.
|
|
||
|
|
||
| ## Normalized Global / Phosphoproteomics Data for Clustering via PCA | ||
|
|
There was a problem hiding this comment.
I feel like the issues below should go in the text of the markdown files themselves, as the markdowns have no explanation or description of results. From here it is not clear which markdown or script I use to generate the figures.
Maybe reduce the text here to point to the markdowns themselves, describe the goal of each file and summarize the figures produced.
|
This is getting closer, but it's still not clear how to reproduce your analysis and conclusions from the code. The readme provides high level overview but doesn't explain how to get there, while the markdowns run the code but don't contextualize the results. The readme should just be telling you what files to run and what their outputs are. The markdowns are the meat of the analysis: why you are running the code, what you are running, and what the results are. |
|
|
||
|
|
||
| # Helpers | ||
|
|
| wide[[sample_col]] <- NULL | ||
| wide | ||
| } | ||
|
|
There was a problem hiding this comment.
at the very minimum comment each function to make it entirely clear what the inputs and outputs are
Modularized the batch correction code, the correlation analysis code, and the leapR code so each could be called with a function call from one of 3 scripts across each of the modalities (RNA, global, phospho).
Also added in the figure and analysis description word doc.