VCF Extract — Genotype Data Helper Scripts

A small collection of fast, standalone Python scripts for processing genotype data from VCF files and related resources (reference FASTA, HPO, ClinVar).

Requirements: Python 3. No heavy dependencies; scripts use the standard library (plus optional network access for ClinVar).

---

Scripts

vcf_to_flanks_fasta.py

Extracts flanking reference sequence for each variant in a VCF.

• Input: VCF + whole-genome FASTA
• Output: FASTA with one record per variant (configurable window, e.g. 100 bp up + ref + 100 bp down)
• Features: Uses a .fai index (built if missing), supports --include-ref, chr1/1 normalization, REF validation

python vcf_to_flanks_fasta.py --fasta genome.fa --vcf variants.vcf --out flanks.fasta
python vcf_to_flanks_fasta.py --fasta genome.fa --vcf variants.vcf --out flanks.fasta --include-ref --up 100 --down 100

---

vcf_clinvar_pathogenic.py

Checks VCF variants against ClinVar via NCBI E-utilities and reports pathogenic / likely pathogenic hits.

• Input: VCF
• Output: Tab-delimited file of variants that are pathogenic or likely pathogenic in ClinVar
• Features: GRCh37 or GRCh38 via --assembly; rate-limited NCBI requests

python vcf_clinvar_pathogenic.py --vcf input.vcf --assembly GRCh38 --out clinvar_hits.tsv

---

hpo_terms_to_gene.py

Predicts likely causal genes from a list of HPO phenotype terms using an HPO annotation database.

• Input: File of HPO term IDs (one per line) + phenotype annotation (HPOA or custom TSV)
• Output: Ranked candidate genes (and optionally top diseases)
• Features: Binary or IC-weighted scoring, optional OBO propagation, disease→gene mapping

python hpo_terms_to_gene.py --hpo_terms patient_hpo.txt --hpoa phenotype.hpoa --topk 20 --show_diseases
python hpo_terms_to_gene.py --hpo_terms terms.txt --tsv disease_hpo_gene.tsv --out results.tsv --weights ic

---

make_example_genome.py

Builds a small example reference FASTA (contigs 16 and 20) with REF bases matching ex2.vcf, for testing vcf_to_flanks_fasta.py without a full genome.

python make_example_genome.py
# Creates genome.fa in the current directory

---

Quick start

1. Clone or download this repo.
2. Run any script with --help for full options, e.g.
   python vcf_to_flanks_fasta.py --help
3. Use ex2.vcf and the genome produced by make_example_genome.py to try the flank extraction pipeline.

---

License

Use and modify as needed for your projects.