Versions.txt

0.1.1
White space in first line of sample list is now removed
Limma/Voom now using R 3.3.2. 
MA and Volcano plots generated with ggplot2 and ggrepel to avoid overlap of labels.
Limma app removes Rplots.pdf and any back up text files
Removed B and t statistic from limma results
0.1.2 ##premature publish
0.1.3
Add back B and t statistic from Limma results
MDS plot
PRocessing of samplesheet with processFiles.py will now exit properly if fastq file not presented in sample sheet.
Also fixed some error catching to print to the error file in processFiles.py
0.1.5
Default billing set to
org-stjude_cloud
GSEA input
Worflow versions to allow set of samples to be sent through pipeline (and more than 20)
sample sheet format changed (Comparisons line needs a '#' at beginning and can be anywhere)
FPKM output
For contrasts with not enough samples a 'simple' differential expression will be performed.  This is done on all samples/contrasts irrespective of LIMMA.  If 3 samples present, MDS will be performed.  Samples are normalized by log2(cpm)

0.2.2
Star version now being used STAR_2.5.3a
Default parameters
STAR --runMode alignReads --genomeDir /home/dnanexus/in/star_index_archive/STARINDEX --readFilesIn readFile1 readFile2          --outFileNamePrefix TEST_LARGE1. --outSAMunmapped Within          --runThreadN 8   --sjdbGTFfile annotation.gtf          --quantMode TranscriptomeSAM --sjdbOverhang 100          --outSAMattributes NH HI AS nM NM MD XS --outFilterMultimapNmax 20 --outFilterMismatchNmax 10          --alignIntronMax 500000      --chimSegmentMin 20 --chimJunctionOverhangMin 20          --outSAMstrandField intronMotif --outBAMcompression -1          --outSAMtype BAM SortedByCoordinate --outBAMsortingThreadN 8          --readFilesCommand zcat

Custom genomes added back.  Now in 'Set Inputs'
Run control to allow certain steps not be used

Log2 FPKM and non log2 FPKM values shown (previously just log2, but was unclear that this was being used)