Skip to content

QUICKSTART.md

Latest commit

 

History

History
97 lines (76 loc) · 2.66 KB

QUICKSTART.md

File metadata and controls

97 lines (76 loc) · 2.66 KB
Raw

MPact Scoring Pipeline - Quick Start

This file is the minimal runbook. Full details are in README.md.

1. Setup

git clone https://github.com/comprna/MPact.git
cd MPact

# First time only
python -m venv .venv
source .venv/bin/activate
pip install -r requirements.txt

2. Bundled paths (all local to this directory)

  • Script: score_mpact.py
  • Recommended model: models/mpact_dtm6a_501nt_seed42.keras
  • Current models: models/mpact_dtm6a_{101,201,501,801,1001}nt_seed42.keras
  • Model metadata and SHA-256 checksums: models/model_manifest.json
  • Required conservation input: a GRCh38 phyloP or phastCons bigWig
  • Accessibility is always calculated with ViennaRNA RNAplfold
  • The public pipeline does not report stoichiometry predictions
  • FASTA: hg38.fa
  • GTF: Homo_sapiens.GRCh38.110.gtf.gz

3. Run on bundled sample (recommended smoke test)

cd MPact
python score_mpact.py \
  --input mini_scoreable.tsv \
  --output-tsv sample_predictions.tsv \
  --fasta hg38.fa \
  --model-path models/mpact_dtm6a_501nt_seed42.keras \
  --window-size 501 \
  --conservation-bigwig /path/to/hg38.phyloP_or_phastCons.bw \
  --output-plot sample_delta_hist.png

4. Run on your own file

TSV input

cd MPact
python score_mpact.py \
  --input /path/to/variants.tsv \
  --output-tsv /path/to/predictions.tsv \
  --fasta hg38.fa \
  --model-path models/mpact_dtm6a_501nt_seed42.keras \
  --window-size 501 \
  --conservation-bigwig /path/to/hg38.phyloP_or_phastCons.bw \
  --output-plot /path/to/delta_histogram.png

VCF input

cd MPact
python score_mpact.py \
  --input /path/to/variants.vcf.gz \
  --output-tsv /path/to/predictions.tsv \
  --fasta hg38.fa \
  --model-path models/mpact_dtm6a_501nt_seed42.keras \
  --window-size 501 \
  --conservation-bigwig /path/to/hg38.phyloP_or_phastCons.bw \
  --gtf Homo_sapiens.GRCh38.110.gtf.gz

5. Required and optional flags

Required:

  • --gtf Homo_sapiens.GRCh38.110.gtf.gz (or custom) — Strand inference from GTF is mandatory for correctness. Bundled default is recommended.

  • --rediportal-gz /path/to/TABLE1_hg38_v3.txt.gz — A-to-I annotations are mandatory.

  • --conservation-bigwig /path/to/track.bw — The track must use the same assembly as the FASTA. Optional:

  • Keep intergenic VCF variants: --allow-nongenic

  • Smaller memory footprint: --batch-size 256

  • Larger neighborhood scan: --scan-radius 20

6. Input requirements

  • Supported input: TSV, VCF, VCF.GZ
  • TSV required columns: #Chromosome, Position, Reference, Alteration
  • SNV input expected (non-SNV rows are dropped)

7. PBS submission

Edit submit_mpact_scoring.pbs, then run:

qsub submit_mpact_scoring.pbs