tooluniverse-variant-analysis

tooluniverse-variant-analysis

Production-ready VCF processing, variant annotation, and mutation analysis for bioinformatics.

Overview

This skill provides comprehensive variant analysis capabilities:

• VCF Parsing: Pure Python + cyvcf2 parsers supporting VCF 4.x, gzipped, multi-sample files
• Mutation Classification: Automatic mapping of SnpEff, VEP, and GATK annotations to standardized mutation types
• Flexible Filtering: VAF, depth, quality, consequence, population frequency, chromosome
• ToolUniverse Integration: Annotation via MyVariant.info (ClinVar, dbSNP, gnomAD, CADD, SIFT, PolyPhen)
• BixBench Support: Designed to answer bioinformatics analysis questions about VCF data

Installation

# Required
pip install pandas

# Recommended (faster VCF parsing)
pip install cyvcf2 pysam

# For ToolUniverse annotation
pip install tooluniverse

Quick Start

from python_implementation import (
    parse_vcf,
    filter_variants,
    compute_variant_statistics,
    answer_vaf_mutation_fraction,
    FilterCriteria,
)

# Parse VCF
vcf_data = parse_vcf("variants.vcf")
print(f"Variants: {len(vcf_data.variants)}, Samples: {vcf_data.samples}")

# Filter
criteria = FilterCriteria(min_vaf=0.1, min_depth=20, pass_only=True)
passing, failing = filter_variants(vcf_data.variants, criteria)

# Statistics
stats = compute_variant_statistics(passing)
print(f"Ti/Tv: {stats['ti_tv_ratio']}")
print(f"Missense: {stats['mutation_types'].get('missense', 0)}")

# Answer BixBench question
result = answer_vaf_mutation_fraction("variants.vcf", max_vaf=0.3, mutation_type="missense")
print(f"Fraction of VAF<0.3 that are missense: {result['fraction']:.4f}")

Files

File	Description
python_implementation.py	Core implementation (parsing, filtering, annotation, reporting)
test_variant_analysis.py	Comprehensive test suite (58 tests)
SKILL.md	Detailed skill documentation
QUICK_START.md	Usage examples for Python SDK and MCP
README.md	This file
test_data/sample.vcf	Test VCF (24 variants, 2 samples, annotated)
test_data/cohort2.vcf	Second test VCF for cohort comparison

Test Results

58/58 tests passed (100%)

Phase 1: VCF Parsing         - 10/10 PASS
Phase 2: Classification       - 10/10 PASS
Phase 3: Filtering            - 12/12 PASS
Phase 4: Statistics           -  4/4  PASS
Phase 5: BixBench Questions   - 10/10 PASS
Phase 6: DataFrame            -  3/3  PASS
Phase 7: Report Generation    -  2/2  PASS
Phase 8: ToolUniverse Annot.  -  3/3  PASS
Phase 9: Edge Cases           -  4/4  PASS

Key Capabilities

Mutation Types Supported

Type	Examples
missense	p.V600E, p.L858R
nonsense	p.R273*
synonymous	p.Ala67=
frameshift	p.Val234Glyfs*6
splice_site	splice_acceptor_variant
splice_region	splice_region_variant
inframe_insertion	p.insAla
inframe_deletion	p.Arg194del
intronic	intron_variant
intergenic	intergenic_variant
UTR_5 / UTR_3	5'/3' UTR variants

VAF Extraction Formats

Format	Field	Computation
Direct AF	FORMAT:AF	Use directly
Allelic Depth	FORMAT:AD	alt / (ref + alt)
FreeBayes	FORMAT:AO/RO	AO / (AO + RO)
NR/NV	FORMAT:NR/NV	NV / NR
INFO AF	INFO:AF	Fallback for all samples

ToolUniverse Annotation Sources

Source	Data	Via
ClinVar	Clinical significance	MyVariant.info
dbSNP	rsID, population frequencies	MyVariant.info or direct
gnomAD	Allele frequencies (global + ancestry)	MyVariant.info or direct
CADD	Deleteriousness score (PHRED)	MyVariant.info
SIFT	Functional prediction	MyVariant.info (CADD)
PolyPhen	Functional prediction	MyVariant.info (CADD)

Running Tests

python test_variant_analysis.py

Command Line Usage

python python_implementation.py variants.vcf -o report.md --annotate --min-vaf 0.1 --pass-only

License

Part of the ToolUniverse project.