The directory contains the supplementary information as well as the software necessary to run the analysis and reproduce figures of the following publication: Katharina Imkeller, Giulia Ambrosi, Nancy Klemm, Ainara Claveras Cabezudo, Luisa Henkel, Wolfgang Huber, Michael Boutros (2021). Metabolic balance in colorectal cancer is maintained by optimal Wnt signaling levels.
The repository is composed by a number of subdirectories, which will be introduced in the following. Please note that due to the large size of the TCGA and DepMap datasets, we do not include the data in the github repositories. You will however find all scripts to assemble the respective datasets in the subfolder data_preparation/.
The external_data/ subfolder contains datasets that were generated in the context of the project and are not published anywhere else. The files depmap_msi_status.txt.gz and zhan_cms.csv.gz contain metainformation on depmap cell lines previously collected by Zhan et al. 2021 (https://doi.org/10.1002/ijc.33393). The file RKO_APC_microarray_resultstable.RData contains the results from the microarray experiment conducted in this study. The file CRISPR_screen_gRNA_counts.csv.gz contains raw gRBNA count data from the CRISPR screens conducted in this study.
In addition, you will have to download the following datasets into the external_data/ folder in order to be able to run the complete analysis.
wget http://linkedomics.org/cptac-colon/Human__CPTAC_COAD__PNNL__Proteome__TMT__03_01_2017__BCM__Gene__Tumor_Normal_log2FC.cct
wget http://linkedomics.org/cptac-colon/Human__CPTAC_COAD__UNC__RNAseq__HiSeq_RNA__03_01_2017__BCM__Gene__BCM_RSEM_UpperQuartile_log2.cct.gz
wget https://www.cell.com/cms/10.1016/j.cell.2019.03.030/attachment/87317e5c-393a-47d6-9c62-cb4fee1d653b/mmc1.xlsx
Extract the B-clinicalData table from mmc1.xlsx as NIHMS1524432-supplement-8.csv.
TCGA_metadata_mutations.Rmd
The script uses the Bioconductor packages TCGAbiolinks and GenomicDataCommons to access RNA expression, mutational and metadata from the TCGA-COAD and TCGA-READ datasets. Raw RNA expression quantification files are downloaded into subdirectories of ./GDCdata/TCGA-READ/ and ./GDCdata/TCGA-COAD/. The corresponding html file including package versions used in the manuscript is available as TCGA_metadata_mutations.html. The metadata and mutation data is stored in the files ../r_data/TCGA_COAD_READ_mutations_mutect.RData and ../r_data/TCGA_COAD_READ_rnaseq_annotation.RData.
TCGA_all_counts_DESeq2.Rmd
The script uses the Bioconductor packages DESeq2 to summarize RNA expression data from the previously downloaded raw expression measures. The corresponding html file including package versions used in the manuscript is available as TCGA_all_counts_DESeq2.html. Summarized normalized and raw RNA sequencing counts are stored in the files ../r_data/TCGA_allcounts_normalized.RData and ../r_data/TCGA_allcounts_raw.RData.
TCGA_classify_CMS_CMSclassifier.Rmd
The script uses the R packages CMSclassifier to classify TCGA-COAD and TCGA-READ samples into consensus molecular subtypes. The corresponding html file including package versions used in the manuscript is available as TCGA_classify_CMS_CMSclassifier.html. CMS classification results are stored in ../r_data/TCGA_CMS_classification_CMSclassifier.RData.
prepare_DepMap_data.Rmd
The script uses the Bioconductor packages depmap to access RNA expression, mutational, gene dependency, proteomic and metadata from the depmap project. The corresponding html file including package versions used in the manuscript is available as prepare_DepMap_data.html. The depmap data concerning colorectal cancer cell lines is stored in ../r_data/depmap_rna_prot_crispr.RData, ../r_data/depmap_metadata.RData and ../r_data/depmap_mutations.RData.
Note that the Differential analysis scripts can only be run after classification of TCGA samples into Wnt-low and Wnt-high, which is implemented in the script main_figures/Figure1.Rmd.
definitions_analysis_visualization.R R script containing variables that are frequently used in other analysis or visualization scripts (thresholds, sizes, colors....).
wnt_low_high_differential_model.Rmd This script implements the quantification of differential transcript and protein expression as well as genetic dependency data in the different datasets. The results are stored in the following files:
TCGA transcriptomics: ../r_data/DESeq2_dds_wnthigh_wntlow_normal.RData, ../r_data/DESeq2_res_high_low.RData, ../r_data/DESeq2_res_high_norm.RData, ../r_data/DESeq2_res_low_norm.RData
TCGA proteomics: ../r_data/tcga_prot_res_high_low.RData
DepMap transcriptomic: ../r_data/depmap_transcriptomics_limma_wntgroup.RData
DepMap proteomic: ../r_data/depmap_proteomics_limma_wntgroup.RData
DepMap gene dependency: ../r_data/depmap_dependency_limma_wntgroup.RData
gsea_functions.R This script contains the common definition of functions utilized for gene set enrichment analysis.
wnt_low_high_gsea_gene_dependencies.Rmd Gene set enrichment analysis for genetic dependency data. This script also generates the subfigures of Figure S3.
wnt_low_high_gsea_transcriptomic_proteomic.Rmd Gene set enrichment analysis for transcriptomic and proteomic data. This script also generates the subfigures of Figure S3.
analyze_CRISPR_screen.Rmd This script quantifies differential gene essentiality in the APCtrunc and APCwt cell lines that were screened in our CRISPR screening experiment.
This folder is used to store all intermediate R objects generated and used in the scripts described above.
The folder contains scripts to generate main figures from the previously performed computational analysis. Figure 5 is not included because it exclusively contains experimental data.
crispr_screen_qc.Rmd Script that produces the supplementary Figure related to CRISPR screen quality control.
supplementary_figures.tex
LaTex file to generate the supplementary_figures.pdf file.