Laytr - Library for variant benchmarking stratification

Install

git clone https://github.com/ACEnglish/laytr.git
cd laytr
python3 -m pip install .

Quick Start

usage: laytr [-h] CMD ...

laytr v0.0.1 Library for variant benchmarking stratification

Available commands:
    kfeat     Create kmer featuration of genomic regions
    map       Map kfeats to a SOM and report neurons
    tru2ga    Consolidate truvari outputs and annotate with GA4GH intermediates
    giabSV06  GIAB SV v0.6 report on a truvari directory
    giabTR    GIAB TR report on a refine.regions.txt

positional arguments:
  CMD         Command to execute
  OPTIONS     Options to pass to the command

Usage

SOMs

Create kmer featurization of genomic regions

laytr kfeat --regions example/example.bed --reference example/reference.fa --output chr22_kfeat.jl

You can then create a SOM out of the kmer featurization. See notebooks/KmerSom.ipynb for an example.

Next, you can map regions to the SOM.

laytr map --input chr22_kfeat.jl --som soms/adotto_TRv1.1_3mers.som --output chr22_kfeat_map.jl

Finally, you can visualize your SOM with notebooks/SOMPlotExample.ipynb.

tru2ga

Consolidate truvari results into GA4GH intermediates.

laytr tru2ga -i truvari_results/ --with-refine -o result_
# Creates `result_truth.vcf.gz` and `result_query.vcf.gz`

giabTR

Creates an html report from truvari's refine.regions.txt on the GIAB TR benchmark.

# Make a truvari result
truvari bench -b giab_tr.vcf.gz -c tr_caller.vcf.gz --includebed giab_tr.bed -o bench/
truvari refine --reference grch38.fa bench/
laytr giabTR -r bench/refine.regions.txt \
	     -b giab_tr.bed \
	     -t adotto_tr_catalog.bed \
	     -s adotto_TRv1.1_3mers.som # available from laytr repo \
	     -m adotto_TRv1.1_3mers.map # available from laytr repo \
	     -o giabTR_report.html

See examples/giabTR_report.html (rendered) for a look at the report.

giabSV06

blank

ToDo

• make SurbScore (no cli)