RNA-seq pipeline, from raw reads to biological interpretaion of time-course experiment. Whenewer possible bash command-line tools are preffered over R.
- English localization
alternatywy.r - just some links to web pages with potentially usefu information.
Directory deprecated
prb.masigpro.r - test of time-course analysis in maSigPro R package.
see.gen2plot.r - modification of the see.genes function, used to prevent over-writing of plots in R (used from command-line) In the end this package will not be used as it don't allow to compare multiple groups and the contact with developers is poor.
Directory test-simulated-data
The first trials of read-processing are given in rnaseq_pipeline.sh
deseq-analysis.r - test of differential gene expression analysis in DESeq2 package in R. Two versions, two replications and three replications
go-enrich.r - GO over and underrepresentation test in goseq
mfuzz-analysis.r - fuzzy (soft) clustering, use data from deseq-analysis.r. Here also corrected script colorbar.R for legend is used.
- other explorative analyses
Directory sequence-analysis rnaseq-ost.sh - the clean read-processing workflow, based on rnaseq_pipeline.sh
rnaseq72.sh - the clean read-processing workflow for whole first replication of experiment
counts4r.sh - construction of files used for DESeqDataSet in DESeq2 package in R
bam4labels.sh - commands used to change header of the counts file to match sample labels
Directory stats-and-visuals
boxplot-r.r - simple boxplot of raw counts
counts-boxplots.r - more advanced boxplots of read counts, involving transformation and FPKM calculation
explorative-plots.r - exploring patterns in count data from first sequencing results (PCA and clustering)
overrepresented-seqs-check.sh - what are overrepresented sequences in FastQC results?
multiplot.r - script for plotting multiple PCA plots (plotPCA command) at the same area
- correct data after new sequencing of "U3" sample - we have results
- full analysis of 2nd replication, plus joint PCA/clustering analysis
- update sample names in count file to include replication number
- check, if it is possible to make DE analysis on two replications (DEseq2) - OK