The QuASAR software package is engineered for joint genotyping of RNA-Seq data, identifying heterozygous loci, and conducting inference on allele specific expression (ASE). Prior to analsyis, RNA-Seq data must undergo alignment with a modern aligner, quality filtering, duplicate removal, and the creation of pileups. There are many tools and tutorials available for preprocessing Next Generation Sequencing data, but we will only describe the tools we used and expect the user to have basic familiarity with standard bioinformatics command-line tools. Our goal with this tutorial is to cover the following:
- Installing QuASAR
- Preprocessing
- Alignment, filtering, and removing duplicates. (Description of, not a tutorial how)
- Pileups and clean pileups
- QuASAR analyis pipeline
- Genotyping single or multiple samples
- Inference on ASE
- Sample workflow
Quick-start: Users comfortable processing RNA-Seq data to the level of pileups should skip to the second step of preprocessing.
require(devtools)
install_github('QuASAR', 'piquelab')
library('QuASAR')From our experience, installing R packages from GitHub within an R session hosted on an ssh conection can be more troublesome than installing on a local machine. An alternative is to clone this repository then build the package:
git clone git@github.com:piquelab/QuASAR.gitzless SomeSample_1.pileup.clean.bed.gz | head -5
chr1 879910 879911 G A rs143853699 0.02 21 0 0
chr1 892379 892380 G A rs150615968 0.0041 22 0 0
chr1 893384 893385 G A rs140972868 0.01 6 0 0
chr1 894101 894102 A T rs188691615 0.01 6 0 0
chr1 894430 894431 G A rs201791495 9e-04 9 0 0ase.joint <- fitAseNull(finalref, finalalt, log.gmat=log(ase.dat.gt$gmat))ase.joint <- fitAseNullMulti(finalref, finalalt, log.gmat=log(ase.dat.gt$gmat))