agentleFS
Sign inStart free
Scientific-ToolingScientific-Tooling/structured-intelligence/skills/genome-variant-calling/SKILL.md

genome-variant-calling

Skill4 starsChanged 12 days agoMITView on GitHub

Call SNVs and indels from aligned BAMs using GATK HaplotypeCaller or DeepVariant with optional GVCF output for joint genotyping.

The file

We split this skill into the rules it gives an agent

3 rules. About 0 are common, 1 is distinctive, and 2 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈1

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 2

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    If GATK: run gatk HaplotypeCaller with --emit-ref-confidence GVCF (if GVCF mode) or standard VCF output.

    line 31
  2. Very specific:

    If DeepVariant: run via Docker or Singularity with the appropriate model type (WGS or WES).

    line 32
  3. Distinctive:

    If intervals are provided (WES): restrict variant calling to target regions with padding.

    line 33

≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Skill: genome-variant-calling

Use When

  • User wants to call germline variants (SNPs, indels) from a deduplicated BAM.
  • User needs to produce a VCF or GVCF for downstream joint genotyping.
  • User wants to compare variant callers (GATK HaplotypeCaller vs DeepVariant).
  • User is performing WGS or WES variant discovery.

Inputs

  • Required:
    • Deduplicated BAM file with index (.bam + .bai).
    • Reference FASTA with .fai index and .dict sequence dictionary.
  • Optional:
    • Caller choice: gatk or deepvariant (default: gatk).
    • Intervals or BED file (for WES target regions).
    • GVCF mode flag (emit reference confidence; default: off).
    • Ploidy (default: 2).
    • Output directory (default: current directory).
    • Threads / memory allocation.

Workflow

  1. Validate that the BAM is sorted, indexed, and has read groups.
  2. If GATK: run gatk HaplotypeCaller with --emit-ref-confidence GVCF (if GVCF mode) or standard VCF output.
  3. If DeepVariant: run via Docker or Singularity with the appropriate model type (WGS or WES).
  4. If intervals are provided (WES): restrict variant calling to target regions with padding.
  5. Index the output VCF/GVCF with bcftools index or gatk IndexFeatureFile.
  6. Report variant counts: total variants, SNPs, indels, het/hom ratio, Ti/Tv ratio.

Output Contract

  • VCF or gVCF file (.vcf.gz).
  • VCF index (.vcf.gz.tbi).
  • Variant summary statistics (total, SNPs, indels, het/hom ratio, Ti/Tv).

Limits

  • GATK requires Java 17+.
  • DeepVariant requires Docker or Singularity (GPU recommended for performance).
  • Joint genotyping (GenomicsDBImport + GenotypeGVCFs) is a separate downstream step.
  • Memory: ~4 GB per thread for GATK HaplotypeCaller; GPU recommended for DeepVariant.
  • This skill handles single-sample calling; multi-sample joint calling is out of scope.
  • Common failure cases:
    • BAM missing read groups, causing GATK HaplotypeCaller to reject the input.
    • Reference FASTA lacking .dict or .fai index files.
    • DeepVariant Docker/Singularity image not found or GPU driver mismatch.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

Check your setup, free