This repository contains the code for my Precision Medicine MSc (University of Leeds) project titled "Investigating therapy-driven changes in isoform expression in glioblastoma".
- 1-dataprocessing - .R scripts for filtering patients and gene/transcripts in the raw in-house and GLASS datasets. Three sub-folders:
- Sample_Filter_Logic - .R scripts for selecting patients to remove based on their metadata and sequencing metric values
- Transcript_Filter_Logic - .R scripts for selecting genes/transcripts to remove based on low expression values
- Filtered_Data - .R scripts to produce final filtered data ready for DEA
- 2-dataexploration - .R scripts for running PCA on in-house gene and in-house/GLASS isoform log2FC data
- 3-dea - .R scripts for running DESeq2 and processing results. Sub-folders:
- in-house - DEA based on in-house data (includes scripts for paired and responder-type DEA, also includes script for DEA based on sub-samples of up-responders)
- glass - DEA based on glass data (includes script for DEA based on sub-sample of up-responders)
- tpm_fpkm_check - per-patient bar plots of TPM (GLASS) and FPKM (in-house) expression values for specific isoforms
- master_deseq2_isoform_results.R - script used to produce Volcano Plots for paired sample DEA results
- 4-resultsanalysis - .R scripts for further exploration of results including:
- GO analysis - .R scripts for extracting lists of interesting genes and visualising overlap in GO terms between different GO enrichment analyses
- summary.dataframe.isoforms.R - summary of all DEA results
- isoform_switch.R - script used to identify cases of isoform switching
- TSS.R - scripts used to map JBSgene transcripts to JARID2 TSSs, and to calculate % of JBS genes with transcripts without a JARID2 TSS
- 5-bashscripts - .sh scripts submitted to the HPC at the University of Leeds
N.B. All folders titled "archive" contain scripts not directly used to produce reported results.