/Intron-reads

Primary LanguageR

Supplementary Figures

Supplementary figure 1 : RNA protocols are compared for gene-level exon log-CPM (left) and gene-level intron log-CPM values (right) for human cell line HCC827 R1, R2, R3, human cell line NCI-H11975 R1, R2, and R3 (in order of plots shown).

Supplementary figure 2 : From left to right, intron versus exon log-CPM, intron versus exon log-RPKM, intron log-RPKM versus total intron length, relative coverage between exon and intron regions versus total intron length, and relative coverage versus exon log-RPKM; with poly(A) RNA (top row) and Total RNA libraries (bottom row). The plots are displayed for human cell line HCC827 R1, R2, R3, human cell line NCI-H11975 R1, R2, and R3 (in order of plots shown).

Scripts for Annotations

gtf_to_exon_saf.R : Convert gtf file to simplified exon locations and save in saf format (also saves GENCODE gene information to txt).

exon_saf_to_genebody_and_intron_saf.R : Exon saf file is used to create genebody saf and intron saf files.

genebody_saf_to_nonoverlapping_genebody_saf.R : Create genebody saf file for non-overlapping genes.

Scripts for Analyses

Exploring intron reads

analysis_of_intron_exploration.R : Main data analysis file that explores basic characteristics of intron reads.

The analysis requires:

  • Annotation files (exon saf and genebody saf)
  • Bam files
  • Sample information (sample names, groups, single/paired-end reads)

The analysis includes:

  • Summarising reads into exon, intron and genebody counts
  • Calculating exon and intron read percentages
  • Creating MDS plots for exon and intron counts
  • Calculating percentage of reads with exon signal and intron signal
  • Calculating correlations for exon and intron log-counts

Characteristics of counts for celllines

analysis_of_characteristics_of_counts_for_celllines.R : Scatter plots for characteristics between and within libraries in cellline data.

Coverage patterns

analysis_of_coverage_for_HCC827.R : Binned coverage analysis for HCC827 human cell line libraries.

The analysis requires:

  • Output from main data analysis
  • Annotation files (exon saf, intron saf, and non-overlapping genes saf)
  • GENCODE gene information
  • Function to create binned annotation: FUN_create_binned_annotation.R
  • Function to plot coverage profiles: FUN_plot_coverage_profile.R

The analysis includes:

  • Selecting a subset of genes to analyse (protein coding and non-overlapping genes in reference chromosomes)
  • Creating binned annotation for exons and introns
  • Summarising reads into exon and intron coverage
  • Examining the distribution of exon and intron coverage
  • Examining the coverage along the genebody for exon and intron regions
  • Creating coverage profile plots for two short and two long genes
  • Creating coverage profile plots of intron retention-like genes
  • Creating coverage profile plots for genes in Rasko paper

IRFinder analysis

Scripts in "IRFinder_analysis" folder.

  • IR_detection.txt : IR detection
  • AC_differential_IR_analysis.txt : Audic and Claverie method for differential IR analysis
  • GLM_differential_IR_analysis.R : GLM method for differential IR analysis
  • filePaths.txt : File paths used in GLM analysis
  • experiment.txt : Experiment information used in GLM analysis

index analysis

index_analyses.R : index analysis of human cell line and immune cells libraries

The analysis requires:

  • DGE objects for introns, exons and genebody for cell lines and immune cells

The analysis includes:

  • Creating barplot for index categories
  • Creating scatter plots for t-statistics and logFC comparisons between introns and exons
  • Creating boxplot distribution of intron lengths for each index category

superintronic analysis

This is a submodule to sa-lee/analysis-superintronic which contains all scripts for elucidating coverage profiles for intron retention signal. It also includes cached data to reproduce the analysis (note this requires git-lfs to be installed).

The analysis requires:

  • IRfinder analysis to be run
  • BAM files to be processed for running coverage
  • fastq files available if running kallisto and isoformSwitchAnalyzeR

This analysis includes:

  • Creating coverage profiles for all IR methods tested in the paper
  • Finding IR signal using superintronic for all cellline x kit combintaions
  • Comparing overlaps between all methods
  • Running isoformSwitchAnalyzeR