Snpsift
DocumentssnpsiftSnpsift is an Agent Skill for documents published in the pcingola/SnpEff repository on GitHub: a SKILL.md instruction file an AI agent can load to do this task.
Run SnpSift, a toolbox for filtering and manipulating annotated VCF files. Use when the user wants to filter variants, annotate with databases (dbNSFP, dbSnp, GWAS), extract fields, or manipulate VCF files.
How to use this skill
Bring this guide into your coding agent with a prompt tailored to the tool you use.
- Open your project in Codex.
- Copy and paste the prompt below.
- Review files and risks before approving any action.
Inspect the Agent Skill at https://github.com/pcingola/SnpEff/blob/HEAD/.claude/skills/snpsift/SKILL.md and all supporting files. Report provenance, permissions, scripts, dependencies, and security risks. Propose an exact installation into .agents/skills/snpsift, but wait for my approval before writing files or running scripts.
Copying this prompt does not install or run the skill. Review third-party files before use.
SnpSift
SnpSift is a toolbox for filtering and manipulating annotated VCF files. Once genomic variants have been annotated, SnpSift helps filter and process them to find relevant variants.
Full documentation: https://pcingola.github.io/SnpEff/
Running
Use the wrapper script at .claude/skills/snpsift/snpsift.sh. It handles JVM memory defaults and argument passthrough. The JAR file is expected at $HOME/snpEff/SnpSift.jar.
IMPORTANT: SnpSift output is typically very large (thousands to millions of lines). ALWAYS redirect output to a file. NEVER let output print to stdout, as it will fill the context window and make the conversation unusable.
# Correct: redirect to file
.claude/skills/snpsift/snpsift.sh <command> [options] [arguments] > output.vcf 2> snpsift.log
# WRONG: never do this
.claude/skills/snpsift/snpsift.sh <command> [options] [arguments]
Commands
| Command | Description | Docs |
|---|---|---|
annotate | Add ID and INFO fields from a VCF database (e.g. dbSnp) | annotate |
annotateMem | Annotate from a VCF database loaded into memory | annotate_mem |
caseControl | Case vs control variant comparison with p-values | casecontrol |
concordance | Concordance metrics between two VCF files | concordance |
dbnsfp | Annotate using dbNSFP (SIFT, Polyphen2, etc.) | dbnsfp |
extractFields | Extract VCF fields to tab-separated format | extractfields |
filter | Filter using arbitrary expressions | filter |
geneSets | Annotate using MSigDb gene sets (GO, KEGG, etc.) | genesets |
gt | Compress genotype fields | gt |
gwasCat | Annotate using GWAS Catalog | gwascatalog |
intersect | Intersect intervals from multiple files | intersect |
intervals | Filter variants by BED intervals | intervals |
intervalsIndex | Filter variants by intervals using file indexing | intervalsindex |
join | Join files by genomic region | join |
phastCons | Annotate using phastCons conservation scores | phastcons |
private | Annotate private variants (family/group) | private |
rmInfo | Remove INFO fields from VCF | rminfo |
rmRefGen | Remove reference genotypes | rmrefgen |
split | Split VCF by chromosome | split |
tstv | Calculate transition/transversion ratio | tstv |
varType | Annotate variant type (SNP, MNP, INS, DEL, MIXED) | varianttype |
vcfCheck | Check VCF file format correctness | vcfcheck |
vcf2tped | Convert VCF to TPED format | vcf2ped |
Additional Documentation
| Topic | File |
|---|---|
| Introduction | introduction |
| FAQ | faq |
| Download & Install | download |
| Examples | examples |
| Help | help |