Skip to content

Latest commit

 

History

History
79 lines (56 loc) · 1.84 KB

File metadata and controls

79 lines (56 loc) · 1.84 KB

phASER Fast Beta - Tester Guide

This bundle contains an optimized phASER build focused on speeding up the default SNP-only workflow (--include_indels 0).

Quick Start (Recommended: conda)

  1. Create the environment:
conda env create -f environment.yml
conda activate phaser-fast-beta
  1. Download the official NA06986 test case (large: BAM is multiple GB):
./scripts/download_test_case.sh
  1. Run the test case (single thread, default MAPQ=255):
./scripts/run_test_case.sh

Outputs go to out/.

Alternative Setup (venv + pip)

If you already have external tools installed (samtools, bcftools, bgzip, tabix, bedtools) and working on your PATH:

python3 -m venv .venv
source .venv/bin/activate
pip install -r requirements.txt

Running On Your Own Data

From the bundle root:

python3 phaser/phaser.py \
  --threads 1 --io_threads 0 \
  --vcf /path/to/sample.vcf.gz \
  --bam /path/to/sample.bam \
  --paired_end 1 \
  --mapq 255 --baseq 10 \
  --sample SAMPLE_ID_IN_VCF \
  --o out/sample

Notes:

  • --mapq 255 is the default used in the upstream NA06986 example.
  • --include_indels 1 is supported but not part of the optimized path in this beta.
  • This beta build is safe to run from paths containing spaces/parentheses (e.g. Dropbox (Personal)).

Smoke Testing Other Tools

This bundle includes:

  • phaser_gene_ae/ (gene-level haplotypic expression)
  • phaser_pop/ (phASER-pop utilities)

You can run:

./scripts/smoke_test.sh

Reporting Issues

Please include:

  • Command used
  • *.log output (phASER writes <prefix>.log)
  • Platform (uname -a) and python version (python3 --version)
  • Whether you used conda or pip

Known non-issue:

  • *.haplotypic_counts.txt may show a different ordering of read-index lists (aReads/bReads) while counts remain the same.