/FSHD-Biopsy-RNA-seq

Preprocessing pipeline and analysis codes for FSHD-Biopsy RNA-seq data

Primary LanguageHTML

This package contains both datasets and scripts (R) supporting the manuscript of MRI-informed muscle biopsies correlate MRI with pathology and DUX4 target gene expression in FSHD. The shell script /scripts/do_tophat.sh performed the preprocessing and /scripts/*.R the analysis for the manuscript. See more detail in /scripts/ReadMe.txt.

Folders

List below are the major datasets and scripts that were used for the analysis. Note that the scripts for data exploratory are also included in the scripts folder, and so they might be a bit messy.

|- data    
  |- se.rda: `RangedSummarizedExperiment` instance as a container containing gene counts, gene annotation and 
     metadata. 
  |- sanitized.dds.rda: sanitied/filtered `DESeqDataSet` of `se`, containing dispersion estimation and size 
     factor, used for downstream analysis  
  |- sanitized.rlg.rda: sanitized/filtered rlog-transformed dataset of sanitized.dds, used to visualize 
     sample space    
  
|- scripts      
  |- *.Rmd:  R markdown document and notebook  
  |- makeSE.R: get gene count and make a `RangedSummarizedExperiment` instance  
  |- rankScore.R: get DUX4 rank and scores
  |- makeManuscripts_scatterPlot.R: make Figure 1

Main analysis

makeSE.R -> rankScore.R -> makeManuscripts_scatterPlot.R -> Figure 1

system requirement

  • R (3.5): pheatmap, xlsx, ggplot2, gridExtra
  • Bioconductor (3.7)
    • DESeqs (differential analysis)
    • goseq (GO analysis)
    • GenomicAlignments (gene counts)
    • org.Hs.eg.db (annotation)
    • GO.db (GO annotation)
  • Optional (Customized pakcages for gene counts)
    • hg38.HomoSapiens.Gencode.v24 (standard TxDb package built from gencode GFF v24)