Run de seq 1 0

analyses/DESeq_analysis/Dockerfile

@@ -0,0 +1,43 @@
+FROM rocker/tidyverse:3.6.0
+# create an R user 
+ENV USER rstudio
+
+
+
+ADD ./analysis/scripts/install_bioc.R install_bioc.R
+
+#ADD ./data/histologies.tsv histologies.tsv
+#ADD ./data/gene-counts-rsem-expected_count-collapsed.rds gene-counts-rsem-expected_count-collapsed.rds
+#ADD ./data/gene-expression-rsem-tpm-collapsed.rds gene-expression-rsem-tpm-collapsed.rds
+#ADD ./data/efo-mondo-map.tsv efo-mondo-map.tsv
+#ADD ./data/uberon-map-gtex-subgroup.tsv uberon-map-gtex-subgroup.tsv
+#ADD ./data/ensg-hugo-rmtl-mapping.tsv ensg-hugo-rmtl-mapping.tsv
+
+ADD ./analysis/run-DESeq-Input-Subsetting.R run-DESeq-Input-Subsetting.R
+ADD ./analysis/run-DESeq-analysis.R run-DESeq-analysis.R 
+#ADD run_deseq.sh run_deseq.sh
+ADD Dockerfile Dockerfile
+
+
+RUN chmod 755 ./install_bioc.R && ./install_bioc.R	DESeq2
+#RUN chmod 755 run_deseq.sh
+
+
+RUN R -e "install.packages('optparse', dependencies = TRUE)"
+RUN R -e "install.packages('jsonlite', dependencies = TRUE)"
+RUN R -e "install.packages('ggplot2', dependencies = TRUE)"
+
+
+#WORKDIR /
+#RUN chmod +x run_deseq.sh
+#RUN ./run_deseq.sh
+
+
+#CMD Rscript --vanilla run-DESeq-Input-Subsetting.R
+#        --hist_file ./data/histologies.tsv \
+#        --counts_file ./data/gene-counts-rsem-expected_count-collapsed.rds \
+#        --tpm_file gene-expression-rsem-tpm-collapsed.rds  \
+#        --efo_mondo_file data/efo-mondo-map.tsv \
+#        --gtex_subgroup_uberon data/uberon-map-gtex-subgroup.tsv \
+#        --ensg_hugo_file data/ensg-hugo-rmtl-mapping.tsv \
+#        --outdir results

analyses/DESeq_analysis/README.md

@@ -0,0 +1,37 @@
+## Differential Expression of RNA-seq matrices
+
+
+This module takes histologies data and the corresponding gene counts and tpm data, and performs differential expression analysis for all combinations of GTEx subgroup and histology type.
+
+
+
+## Expected Input
+
+Data files must be downloaded using the below script
+```
+download-data.sh
+```
+
+
+
+
+## Scripts 
+
+`run-DESeq-analysis.R` - This is the main script that reads the downloaded data files and prints out json and tsv files.
+
+`run_deseq.sh` - This script sets the path for the output tables and calls the above R script.
+
+`process_test_input.R` - This script will create a subset of histologies.tsv. The script searches the cancer_group and cohort combinations that have more than 5 patients with clinical data, and requires the user to specify the desired cancer_group and cohort to use for the new subset creation.
+
+`run_process_test_input.sh` - This script runs the create_test_input.R. This works well with v7 data since the selected cancer_group and cohort to subset for, are reviewed to have more than 5 patients with clinical data. This script will use the newly created subset and process it by running run_DESeq_analysis.R, and create test result table.
+
+
+
+
+## Steps
+1) Set working directory to /deseq_analysis
+2) For testing with v7, run the script run_process_test_input.sh . 
+3) Review the path and file created as test data, and the subsequent result tables created
+4) In order to run the module for the entire set of data, run run_deseq.sh
+
+