Skip to content

Code and figures related to Groves et al. submitted to Cell Systems 2021

Notifications You must be signed in to change notification settings

QuLab-VU/Groves-CellSys2022

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

89 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Groves-CellSys2022

Code and figures related to Groves et al. Cell Systems 2022

Repository Structure:

  • data
    • bulk-rna-seq : contains all data related to bulk RNA seq analyses, used and generated in notebooks/bulk and notebooks/ParTI-code.
    • single-cell-rna-seq
  • mazebox : this can be used like a python package and contains several functions used in notebooks/single-cell.
    • archetypes
    • plasticity
    • plotting
    • preprocessing
    • __init__.py
    • _settings.py
  • notebooks
    • out : output from notebooks, including a Batch Correction report for combining bulk RNA seq from cell lines and tumors.
    • ParTI-code : MATLAB code to run archetype analysis on bulk RNA seq data.
      • human-cell-lines
        • out : contains original five archetypes shown in paper, as well as other numbers of archetypes tested.
        • Human-cell-line-RNA-seq.mlx : Code to generate archetypes for human cell line bulk RNA-seq.
        • params.txt : table showing parameters for each ParTI run, including barcode used in file structure in out
      • thomas-tumors : contains code and results from ParTI for different numbers of archetypes on bulk RNA-seq of 81 human tumors. Same file structure as human-cell-lines.
      • combined-data : contains code and results from ParTI for different numbers of archetypes on combined bulk RNA-seq dataset from notebooks/bulk/Cell-line-tumor-batch-correction-and-clustering.Rmd, using the file data/bulk-rna-seq/parti-input/CCLE_Minna_Thomas_COMBAT_no_names.csv. Same file structure as human-cell-lines.
    • bulk
      • int : contains intermediary files that were saved for antiquity but are not used in any scripts.
      • Bulk-Archetype-Tasks.ipynb : plots related to the inferred tasks for each archetype. This analysis shows N is enriched in axonogenesis genes and migration, A may correspond to TPCs, and Y corresponds to transit-amplifying PNECs.
      • SCLC RNA-seq batch correction-CCLE-Minna.Rmd : batch corrects CCLE and Minna data, as used in the manuscript.
      • Cell-line-tumor-batch-correction-and-clustering.Rmd : batch corrects CCLE, Minna, and Tumor datasets. This analysis is used for ParTI on cell lines and tumors combined.
      • Compare_cell_lines_tumors_archetypes.Rmd : comparing cell line archetypes to combined dataset archetypes.
      • Thomas-Tumors-Bulk-Archetypes.ipynb : comparing the bulk RNA-seq from 81 human tumors to the original five archetypes in the manuscript (from cell line data). This analysis shows that tumor samples are encompassed within the cell line archetype space, and tumor data alone shows more variation in ribosomal proteins than cell lines which may not be phenotypically relevant.
    • single-cell
      • Human_cell_lines.ipynb
      • RPM_tumors.ipynb
      • RPM_tumors-CR.ipynb: using CellRank on RPM time series and tumor data
  • ParTI : ParTI package code, which is used in notebooks/ParTI-code.
  • R_code : R code from Darren Tyson
  • environment.yml : conda environment

Binder