nRex is a set of script to call germline and somatic single-nucleotide variants, short indels and structural variants. The workflow is primarily for human whole-genome sequencing data (>=30x) using short or long-reads.
git clone --recursive https://github.com/tobiasrausch/nRex.git
cd nRex/
make all
cd genome/ && ./prepare_genome.sh
cd refpanel/ && ./download.sh
cd maps/ && ./download.sh
To activate the annotation of variants using VEP, you need to download an offline annotation cache for GRCh38
make .vep
There is a pipeline for short- and long-reads. For short-reads:
./src/nRex.sh <output prefix> <sample1.read1.fq.gz> <sample1.read2.fq.gz>
and for long-reads:
./src/nRexLR.sh <output prefix> <sample.fq.gz>
A few helper scripts to summarize the output of the various tools.
./src/aggregate.sh table *.qc.summary
To generate a table of annotated variants, you can use
python3 scripts/vep.py -v sample.vep.bcf