[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M6DAWXAERYBNEWZH3R6CS0CD":3,"skill-same:01M6DAWXAERYBNEWZH3R6CS0CD":58},{"skillNo":4,"name":5,"nameZh":6,"description":7,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":16,"categories":17,"keywords":18,"publishedAt":33,"linkOnly":34,"markdownContent":35,"markdownBody":36,"frontMatter":37,"refName":38,"commitSha":39,"skillDirectory":40,"forksCount":14,"repositoryHtmlUrl":41,"repositoryHomepageUrl":6,"ownerHtmlUrl":42,"ownerBlogUrl":6,"ownerDescription":6,"updatedAt":43,"install":44,"resources":57},"01M6DAWXAERYBNEWZH3R6CS0CD","tooluniverse-regulatory-variant-analysis",null,"Regulatory variant interpretation -- GWAS association lookup, eQTL analysis, chromatin state annotation, regulatory element overlap, and trait ontology resolution. Connects GWAS Catalog, GTEx, ENCODE, RegulomeDB, OpenTargets, OLS ontology, and Ensembl regulatory features. Use when users ask about non-coding variants, GWAS hits, eQTLs, regulatory elements, enhancer\u002Fpromoter variants, or trait-associated SNPs.","Prekzursil\u002Fagent-skills-toolchain","Prekzursil","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F54636077?v=4","01M6DACZ1PCQHD99SCR1YXG28B","Andrei Mihai Visalon","01M6DACZ670WZF535V3WYYSS4Z",0,"MIT","https:\u002F\u002Fgithub.com\u002FPrekzursil\u002Fagent-skills-toolchain\u002Fblob\u002Fmain\u002F.agents\u002Flibrary\u002Fskills\u002Ftooluniverse-regulatory-variant-analysis\u002FSKILL.md",[],[19,20,21,22,23,24,25,26,27,28,29,30,31,32],"analysis","variant analysis","regulatory","tooluniverse regulatory variant analysis","interpretation","state","lookup","coding","ask","catalog","use","ontology","resolution","elements","2026-11-10 10:12:30",false,"---\nname: tooluniverse-regulatory-variant-analysis\ndescription: Regulatory variant interpretation -- GWAS association lookup, eQTL analysis, chromatin state annotation, regulatory element overlap, and trait ontology resolution. Connects GWAS Catalog, GTEx, ENCODE, RegulomeDB, OpenTargets, OLS ontology, and Ensembl regulatory features. Use when users ask about non-coding variants, GWAS hits, eQTLs, regulatory elements, enhancer\u002Fpromoter variants, or trait-associated SNPs.\n---\n\n## COMPUTE, DON'T DESCRIBE\nWhen analysis requires computation (statistics, data processing, scoring, enrichment), write and run Python code via Bash. Don't describe what you would do — execute it and report actual results. Use ToolUniverse tools to retrieve data, then Python (pandas, scipy, statsmodels, matplotlib) to analyze it.\n\n# Regulatory Variant Analysis Skill\n\nSystematic regulatory variant interpretation: discover trait associations from GWAS, map eQTL effects, annotate chromatin context, assess regulatory element overlap, and produce evidence-graded functional impact predictions for non-coding variants.\n\n## When to Use\n\n- \"What GWAS associations exist for rs12913832?\"\n- \"Find eQTLs for the APOE locus in brain tissue\"\n- \"What regulatory elements overlap this variant region?\"\n- \"Which SNPs are associated with type 2 diabetes from GWAS?\"\n- \"Is this intronic variant in an active enhancer?\"\n- \"What is the RegulomeDB score for rs429358?\"\n- \"Find ENCODE histone marks at the BRCA1 promoter region\"\n- \"Map trait ontology terms for 'blood pressure' to EFO IDs\"\n\n**NOT for** (use other skills instead):\n- Coding variant pathogenicity -> Use `tooluniverse-variant-interpretation`\n- Full clinical variant classification (ACMG) -> Use `tooluniverse-variant-interpretation`\n- Gene-disease associations (not variant-specific) -> Use `tooluniverse-gene-disease-association`\n- Pharmacogenomic variant annotation -> Use `tooluniverse-pharmacogenomics`\n- Epigenomics data processing (BED\u002FnarrowPeak files) -> Use `tooluniverse-epigenomics`\n\n---\n\n## Non-Coding Variant Impact Reasoning\n\nWhen evaluating a non-coding variant, build evidence across four questions:\n\n**1. Is the variant in a regulatory element?**\nUse RegulomeDB to assess whether the variant overlaps TF binding sites, chromatin accessibility peaks, or known regulatory annotations. A low RegulomeDB score (categories 1a-2a) indicates strong evidence that the position is functionally active. Confirm with ENCODE histone marks: H3K27ac signals active enhancers and active promoters; H3K4me1 alone marks poised enhancers; H3K4me3 marks active promoters; H3K27me3 marks silenced regions.\n\n**2. Does it alter a transcription factor binding site?**\nCheck RegulomeDB's TF binding evidence and ENCODE TF ChIP-seq experiments. A variant that falls within a TF footprint and disrupts the consensus motif is mechanistically actionable, especially if the TF is known to be relevant in the disease tissue.\n\n**3. Is there eQTL evidence linking it to a gene?**\nQuery GTEx to determine whether the variant (or variants in tight LD) modulates expression of a nearby gene in a tissue-specific or ubiquitous manner. A tissue-specific eQTL suggests cell-type-specific regulation; a ubiquitous eQTL suggests a core regulatory element. The direction of the NES (positive = alternative allele increases expression, negative = decreases) and effect size matter for interpretation.\n\n**4. Is there GWAS evidence for trait association?**\nSearch the GWAS Catalog for the rsID or the surrounding locus. Genome-wide significant associations (p \u003C 5×10⁻⁸) in relevant traits anchor the variant's biological importance. Cross-reference with OpenTargets for locus-to-gene mapping from multiple GWAS studies.\n\n**Synthesizing the evidence**: Build a multi-layer case. A variant with GWAS significance + eQTL evidence + RegulomeDB score 1a-2a + active chromatin (H3K27ac) in the relevant tissue represents high-confidence regulatory impact. Two or three converging lines of evidence (e.g., eQTL plus active enhancer) constitute moderate confidence. A single line, or a variant only in a poised but not active regulatory context, represents lower confidence.\n\n---\n\n## Workflow Overview\n\n```\nInput (rsID, genomic coordinates, trait\u002Fdisease, gene)\n  |\n  v\nPhase 0: Variant\u002FTrait Resolution\n  Resolve rsIDs, map trait names to EFO\u002FMONDO IDs via OLS\n  |\n  v\nPhase 1: GWAS Association Lookup\n  GWAS Catalog associations, p-values, effect sizes, study metadata\n  |\n  v\nPhase 2: eQTL Analysis\n  GTEx tissue-specific eQTLs, target gene identification\n  |\n  v\nPhase 3: Regulatory Element Annotation\n  ENCODE histone marks, RegulomeDB scores, chromatin state\n  |\n  v\nPhase 4: OpenTargets GWAS Integration\n  OpenTargets GWAS study aggregation, locus-to-gene mapping\n  |\n  v\nPhase 5: Functional Impact Synthesis\n  Integrate all evidence, assign regulatory impact level\n  |\n  v\nPhase 6: Report\n  Evidence-graded regulatory variant report\n```\n\n---\n\n## Phase 0: Variant\u002FTrait Resolution\n\nUse `ols_search_terms` to resolve trait names to ontology IDs before GWAS queries. Restrict to `ontology=\"efo\"` for GWAS traits; OpenTargets prefers MONDO IDs (e.g., MONDO_0005148 for type 2 diabetes rather than EFO_0001360). Use `EnsemblVEP_annotate_rsid` (param is `variant_id`, not `rsid`) for initial consequence annotation and nearest gene identification.\n\n---\n\n## Phase 1: GWAS Association Lookup\n\n`gwas_search_associations` is the primary tool: accepts `disease_trait` (free text), `efo_id` (preferred for precision), `rs_id`, and `p_value` threshold. Use `p_value=5e-8` for genome-wide significance. For locus-level discovery, `gwas_get_variants_for_trait` retrieves all SNPs for a trait. `gwas_get_snps_for_gene` finds GWAS-cataloged SNPs mapped to a specific gene.\n\n**Reasoning tip**: When GWAS Catalog returns empty for a free-text trait, switch to the `efo_id` parameter — the catalog uses controlled vocabulary and free-text matching is imprecise.\n\n---\n\n## Phase 2: eQTL Analysis\n\n`GTEx_query_eqtl` accepts a gene symbol (auto-resolved to GENCODE ID) or Ensembl gene ID. It returns tissue-specific SNP-gene associations with NES (normalized effect size) and p-value per tissue.\n\nWhen interpreting results, ask: does the eQTL effect occur in the tissue most relevant to the disease? A brain-specific eQTL for a neurodegenerative disease variant is more compelling than a ubiquitous one. Use `GTEx_get_median_gene_expression` to confirm that the target gene is actually expressed in the relevant tissue before placing weight on eQTL evidence.\n\n**Note**: GTEx API uses v8 data; gtex_v10 endpoints may return empty for some queries.\n\n---\n\n## Phase 3: Regulatory Element Annotation\n\n`RegulomeDB_query_variant` (param: `rsid`) returns a regulatory score and feature annotations. Scores in categories 1a–2a indicate strong regulatory evidence (eQTL overlap + TF binding + chromatin accessibility). Scores 3a–6 represent progressively weaker evidence.\n\n`ENCODE_search_histone_experiments` accepts `histone_mark` (e.g., \"H3K27ac\") and `biosample_term_name` (tissue or cell line name — NOT a disease name; ENCODE uses biological sample names like \"liver\" or \"breast epithelium\"). Use `assay_title=\"TF ChIP-seq\"` (not just \"ChIP-seq\") when querying TF binding data.\n\n**Reasoning tip**: RegulomeDB aggregates ENCODE, Roadmap, and other data. If ENCODE doesn't have the specific biosample, RegulomeDB may still have aggregate evidence from related cell types.\n\n---\n\n## Phase 4: OpenTargets GWAS Integration\n\n`OpenTargets_search_gwas_studies_by_disease` takes `diseaseIds` as an array of MONDO IDs. It provides locus-to-gene (L2G) scores from multiple GWAS studies, which go beyond simple proximity to incorporate colocalisation, eQTL, and chromatin data. Use `OpenTargets_multi_entity_search` or `OpenTargets_get_disease_id_description_by_name` to resolve disease names to MONDO\u002FEFO IDs first.\n\n---\n\n## Phase 5: Functional Impact Synthesis\n\nAfter collecting evidence, reason through the layers:\n\n- **High impact**: GWAS genome-wide significant + eQTL with meaningful NES + RegulomeDB score ≤ 2 + active chromatin (H3K27ac) in relevant tissue. Multiple independent lines converge on the same locus and gene.\n- **Moderate impact**: Two to three lines of evidence (e.g., eQTL + active enhancer overlap, or GWAS significant + RegulomeDB ≤ 3) without full convergence.\n- **Low impact**: Single line of evidence, or only computational annotation (VEP consequence category) without functional data.\n- **No evidence**: No regulatory annotations in any source; the variant may be in a non-functional region or the relevant cell type is not represented in available datasets.\n\n---\n\n## Fallback Strategies\n\n- **GWAS Catalog returns empty**: Switch from free-text `disease_trait` to `efo_id`; broaden the trait term.\n- **GTEx eQTL empty for gene**: Verify gene symbol spelling; try Ensembl ID; increase `size` parameter.\n- **RegulomeDB returns no data**: Query ENCODE directly; the variant may lack regulatory annotations in available data.\n- **OpenTargets GWAS returns None**: Verify MONDO\u002FEFO ID format; try `OpenTargets_multi_entity_search` first to confirm the correct ID.\n- **ENCODE tissue not found**: ENCODE uses specific biosample names; RegulomeDB aggregates data from many cell types and may cover the gap.\n\n---\n\n## Example Workflows\n\n### GWAS Variant Functional Annotation (rs429358 \u002F APOE)\n\n```\nStep 1: gwas_search_associations(rs_id=\"rs429358\")\n  -> All trait associations (Alzheimer's disease, LDL cholesterol, etc.)\n\nStep 2: GTEx_query_eqtl(gene_symbol=\"APOE\")\n  -> Tissue-specific eQTL evidence; note effect in brain vs liver\n\nStep 3: RegulomeDB_query_variant(rsid=\"rs429358\")\n  -> Regulatory score and TF binding annotations\n\nStep 4: ENCODE_search_histone_experiments(histone_mark=\"H3K27ac\", biosample_term_name=\"brain\")\n  -> Active enhancer context near the variant\n\nStep 5: Synthesize: does GWAS significance + eQTL + active chromatin converge on one gene?\n```\n\n### Non-Coding Variant Assessment (Intronic\u002FUTR Variant)\n\n```\nStep 1: EnsemblVEP_annotate_rsid(variant_id=\"rs12345678\")\n  -> Confirm non-coding consequence, identify nearest gene\n\nStep 2: RegulomeDB_query_variant(rsid=\"rs12345678\")\n  -> Is this position in a regulatory context?\n\nStep 3: gwas_search_associations(rs_id=\"rs12345678\")\n  -> Any GWAS associations in relevant traits?\n\nStep 4: GTEx_query_eqtl(gene_symbol=nearest_gene)\n  -> Does this variant or nearby variants modulate expression?\n\nStep 5: ENCODE_search_histone_experiments(histone_mark=\"H3K27ac\", biosample_term_name=relevant_tissue)\n  -> Active chromatin confirmation\n\nStep 6: Classify impact based on convergence of evidence lines\n```\n\n---\n\n## Limitations\n\n- GWAS Catalog covers published GWAS only; unpublished studies are not included.\n- GTEx eQTL data is from v8; v10 endpoints may return empty.\n- RegulomeDB annotations depend on available ENCODE\u002FRoadmap data for the specific cell type.\n- eQTL analysis identifies correlation, not causation; fine-mapping is needed to identify causal variants.\n- RegulomeDB scores are heuristic; a score of 1a does not guarantee functional impact.\n- GWAS associations are population-level; individual variant effects depend on genetic background.\n","## COMPUTE, DON'T DESCRIBE\nWhen analysis requires computation (statistics, data processing, scoring, enrichment), write and run Python code via Bash. Don't describe what you would do — execute it and report actual results. Use ToolUniverse tools to retrieve data, then Python (pandas, scipy, statsmodels, matplotlib) to analyze it.\n\n# Regulatory Variant Analysis Skill\n\nSystematic regulatory variant interpretation: discover trait associations from GWAS, map eQTL effects, annotate chromatin context, assess regulatory element overlap, and produce evidence-graded functional impact predictions for non-coding variants.\n\n## When to Use\n\n- \"What GWAS associations exist for rs12913832?\"\n- \"Find eQTLs for the APOE locus in brain tissue\"\n- \"What regulatory elements overlap this variant region?\"\n- \"Which SNPs are associated with type 2 diabetes from GWAS?\"\n- \"Is this intronic variant in an active enhancer?\"\n- \"What is the RegulomeDB score for rs429358?\"\n- \"Find ENCODE histone marks at the BRCA1 promoter region\"\n- \"Map trait ontology terms for 'blood pressure' to EFO IDs\"\n\n**NOT for** (use other skills instead):\n- Coding variant pathogenicity -> Use `tooluniverse-variant-interpretation`\n- Full clinical variant classification (ACMG) -> Use `tooluniverse-variant-interpretation`\n- Gene-disease associations (not variant-specific) -> Use `tooluniverse-gene-disease-association`\n- Pharmacogenomic variant annotation -> Use `tooluniverse-pharmacogenomics`\n- Epigenomics data processing (BED\u002FnarrowPeak files) -> Use `tooluniverse-epigenomics`\n\n---\n\n## Non-Coding Variant Impact Reasoning\n\nWhen evaluating a non-coding variant, build evidence across four questions:\n\n**1. Is the variant in a regulatory element?**\nUse RegulomeDB to assess whether the variant overlaps TF binding sites, chromatin accessibility peaks, or known regulatory annotations. A low RegulomeDB score (categories 1a-2a) indicates strong evidence that the position is functionally active. Confirm with ENCODE histone marks: H3K27ac signals active enhancers and active promoters; H3K4me1 alone marks poised enhancers; H3K4me3 marks active promoters; H3K27me3 marks silenced regions.\n\n**2. Does it alter a transcription factor binding site?**\nCheck RegulomeDB's TF binding evidence and ENCODE TF ChIP-seq experiments. A variant that falls within a TF footprint and disrupts the consensus motif is mechanistically actionable, especially if the TF is known to be relevant in the disease tissue.\n\n**3. Is there eQTL evidence linking it to a gene?**\nQuery GTEx to determine whether the variant (or variants in tight LD) modulates expression of a nearby gene in a tissue-specific or ubiquitous manner. A tissue-specific eQTL suggests cell-type-specific regulation; a ubiquitous eQTL suggests a core regulatory element. The direction of the NES (positive = alternative allele increases expression, negative = decreases) and effect size matter for interpretation.\n\n**4. Is there GWAS evidence for trait association?**\nSearch the GWAS Catalog for the rsID or the surrounding locus. Genome-wide significant associations (p \u003C 5×10⁻⁸) in relevant traits anchor the variant's biological importance. Cross-reference with OpenTargets for locus-to-gene mapping from multiple GWAS studies.\n\n**Synthesizing the evidence**: Build a multi-layer case. A variant with GWAS significance + eQTL evidence + RegulomeDB score 1a-2a + active chromatin (H3K27ac) in the relevant tissue represents high-confidence regulatory impact. Two or three converging lines of evidence (e.g., eQTL plus active enhancer) constitute moderate confidence. A single line, or a variant only in a poised but not active regulatory context, represents lower confidence.\n\n---\n\n## Workflow Overview\n\n```\nInput (rsID, genomic coordinates, trait\u002Fdisease, gene)\n  |\n  v\nPhase 0: Variant\u002FTrait Resolution\n  Resolve rsIDs, map trait names to EFO\u002FMONDO IDs via OLS\n  |\n  v\nPhase 1: GWAS Association Lookup\n  GWAS Catalog associations, p-values, effect sizes, study metadata\n  |\n  v\nPhase 2: eQTL Analysis\n  GTEx tissue-specific eQTLs, target gene identification\n  |\n  v\nPhase 3: Regulatory Element Annotation\n  ENCODE histone marks, RegulomeDB scores, chromatin state\n  |\n  v\nPhase 4: OpenTargets GWAS Integration\n  OpenTargets GWAS study aggregation, locus-to-gene mapping\n  |\n  v\nPhase 5: Functional Impact Synthesis\n  Integrate all evidence, assign regulatory impact level\n  |\n  v\nPhase 6: Report\n  Evidence-graded regulatory variant report\n```\n\n---\n\n## Phase 0: Variant\u002FTrait Resolution\n\nUse `ols_search_terms` to resolve trait names to ontology IDs before GWAS queries. Restrict to `ontology=\"efo\"` for GWAS traits; OpenTargets prefers MONDO IDs (e.g., MONDO_0005148 for type 2 diabetes rather than EFO_0001360). Use `EnsemblVEP_annotate_rsid` (param is `variant_id`, not `rsid`) for initial consequence annotation and nearest gene identification.\n\n---\n\n## Phase 1: GWAS Association Lookup\n\n`gwas_search_associations` is the primary tool: accepts `disease_trait` (free text), `efo_id` (preferred for precision), `rs_id`, and `p_value` threshold. Use `p_value=5e-8` for genome-wide significance. For locus-level discovery, `gwas_get_variants_for_trait` retrieves all SNPs for a trait. `gwas_get_snps_for_gene` finds GWAS-cataloged SNPs mapped to a specific gene.\n\n**Reasoning tip**: When GWAS Catalog returns empty for a free-text trait, switch to the `efo_id` parameter — the catalog uses controlled vocabulary and free-text matching is imprecise.\n\n---\n\n## Phase 2: eQTL Analysis\n\n`GTEx_query_eqtl` accepts a gene symbol (auto-resolved to GENCODE ID) or Ensembl gene ID. It returns tissue-specific SNP-gene associations with NES (normalized effect size) and p-value per tissue.\n\nWhen interpreting results, ask: does the eQTL effect occur in the tissue most relevant to the disease? A brain-specific eQTL for a neurodegenerative disease variant is more compelling than a ubiquitous one. Use `GTEx_get_median_gene_expression` to confirm that the target gene is actually expressed in the relevant tissue before placing weight on eQTL evidence.\n\n**Note**: GTEx API uses v8 data; gtex_v10 endpoints may return empty for some queries.\n\n---\n\n## Phase 3: Regulatory Element Annotation\n\n`RegulomeDB_query_variant` (param: `rsid`) returns a regulatory score and feature annotations. Scores in categories 1a–2a indicate strong regulatory evidence (eQTL overlap + TF binding + chromatin accessibility). Scores 3a–6 represent progressively weaker evidence.\n\n`ENCODE_search_histone_experiments` accepts `histone_mark` (e.g., \"H3K27ac\") and `biosample_term_name` (tissue or cell line name — NOT a disease name; ENCODE uses biological sample names like \"liver\" or \"breast epithelium\"). Use `assay_title=\"TF ChIP-seq\"` (not just \"ChIP-seq\") when querying TF binding data.\n\n**Reasoning tip**: RegulomeDB aggregates ENCODE, Roadmap, and other data. If ENCODE doesn't have the specific biosample, RegulomeDB may still have aggregate evidence from related cell types.\n\n---\n\n## Phase 4: OpenTargets GWAS Integration\n\n`OpenTargets_search_gwas_studies_by_disease` takes `diseaseIds` as an array of MONDO IDs. It provides locus-to-gene (L2G) scores from multiple GWAS studies, which go beyond simple proximity to incorporate colocalisation, eQTL, and chromatin data. Use `OpenTargets_multi_entity_search` or `OpenTargets_get_disease_id_description_by_name` to resolve disease names to MONDO\u002FEFO IDs first.\n\n---\n\n## Phase 5: Functional Impact Synthesis\n\nAfter collecting evidence, reason through the layers:\n\n- **High impact**: GWAS genome-wide significant + eQTL with meaningful NES + RegulomeDB score ≤ 2 + active chromatin (H3K27ac) in relevant tissue. Multiple independent lines converge on the same locus and gene.\n- **Moderate impact**: Two to three lines of evidence (e.g., eQTL + active enhancer overlap, or GWAS significant + RegulomeDB ≤ 3) without full convergence.\n- **Low impact**: Single line of evidence, or only computational annotation (VEP consequence category) without functional data.\n- **No evidence**: No regulatory annotations in any source; the variant may be in a non-functional region or the relevant cell type is not represented in available datasets.\n\n---\n\n## Fallback Strategies\n\n- **GWAS Catalog returns empty**: Switch from free-text `disease_trait` to `efo_id`; broaden the trait term.\n- **GTEx eQTL empty for gene**: Verify gene symbol spelling; try Ensembl ID; increase `size` parameter.\n- **RegulomeDB returns no data**: Query ENCODE directly; the variant may lack regulatory annotations in available data.\n- **OpenTargets GWAS returns None**: Verify MONDO\u002FEFO ID format; try `OpenTargets_multi_entity_search` first to confirm the correct ID.\n- **ENCODE tissue not found**: ENCODE uses specific biosample names; RegulomeDB aggregates data from many cell types and may cover the gap.\n\n---\n\n## Example Workflows\n\n### GWAS Variant Functional Annotation (rs429358 \u002F APOE)\n\n```\nStep 1: gwas_search_associations(rs_id=\"rs429358\")\n  -> All trait associations (Alzheimer's disease, LDL cholesterol, etc.)\n\nStep 2: GTEx_query_eqtl(gene_symbol=\"APOE\")\n  -> Tissue-specific eQTL evidence; note effect in brain vs liver\n\nStep 3: RegulomeDB_query_variant(rsid=\"rs429358\")\n  -> Regulatory score and TF binding annotations\n\nStep 4: ENCODE_search_histone_experiments(histone_mark=\"H3K27ac\", biosample_term_name=\"brain\")\n  -> Active enhancer context near the variant\n\nStep 5: Synthesize: does GWAS significance + eQTL + active chromatin converge on one gene?\n```\n\n### Non-Coding Variant Assessment (Intronic\u002FUTR Variant)\n\n```\nStep 1: EnsemblVEP_annotate_rsid(variant_id=\"rs12345678\")\n  -> Confirm non-coding consequence, identify nearest gene\n\nStep 2: RegulomeDB_query_variant(rsid=\"rs12345678\")\n  -> Is this position in a regulatory context?\n\nStep 3: gwas_search_associations(rs_id=\"rs12345678\")\n  -> Any GWAS associations in relevant traits?\n\nStep 4: GTEx_query_eqtl(gene_symbol=nearest_gene)\n  -> Does this variant or nearby variants modulate expression?\n\nStep 5: ENCODE_search_histone_experiments(histone_mark=\"H3K27ac\", biosample_term_name=relevant_tissue)\n  -> Active chromatin confirmation\n\nStep 6: Classify impact based on convergence of evidence lines\n```\n\n---\n\n## Limitations\n\n- GWAS Catalog covers published GWAS only; unpublished studies are not included.\n- GTEx eQTL data is from v8; v10 endpoints may return empty.\n- RegulomeDB annotations depend on available ENCODE\u002FRoadmap data for the specific cell type.\n- eQTL analysis identifies correlation, not causation; fine-mapping is needed to identify causal variants.\n- RegulomeDB scores are heuristic; a score of 1a does not guarantee functional impact.\n- GWAS associations are population-level; individual variant effects depend on genetic background.\n",{"name":5,"description":7},"main","a298c929b6cbd662e1fce54302ce520d5ceea1a5",".agents\u002Flibrary\u002Fskills\u002Ftooluniverse-regulatory-variant-analysis","https:\u002F\u002Fgithub.com\u002FPrekzursil\u002Fagent-skills-toolchain","https:\u002F\u002Fgithub.com\u002FPrekzursil","2026-11-02 14:30:57",{"source":8,"skillName":5,"commands":45,"prompt":55,"zipAvailable":56},[46,49,52],{"tool":47,"command":48},"npx","npx skills add Prekzursil\u002Fagent-skills-toolchain --skill tooluniverse-regulatory-variant-analysis",{"tool":50,"command":51},"bunx","bunx skills add Prekzursil\u002Fagent-skills-toolchain --skill tooluniverse-regulatory-variant-analysis",{"tool":53,"command":54},"pnpm","pnpm dlx skills add Prekzursil\u002Fagent-skills-toolchain --skill tooluniverse-regulatory-variant-analysis","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002FPrekzursil\u002Fagent-skills-toolchain\u002Fblob\u002Fmain\u002F.agents\u002Flibrary\u002Fskills\u002Ftooluniverse-regulatory-variant-analysis\u002FSKILL.md",true,[],[59,80,111,138,151,167],{"skillNo":60,"name":61,"nameZh":6,"description":62,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":63,"categories":64,"keywords":65,"publishedAt":33},"01M6DF7D7478R55P46REYH4QAC","clarify-requirements","Resolve genuine pre-execution ambiguity by scanning repo context, generating 1-6 grounded options, and asking 1-3 focused questions via AskUserQuestion. NOT for mid-task use (injection risk); NOT when the goal is already unambiguous; use to disambiguate vague prompts before any file edits.","https:\u002F\u002Fgithub.com\u002FPrekzursil\u002Fagent-skills-toolchain\u002Fblob\u002Fmain\u002Ftools\u002Fworkspace-hooks\u002Fskills\u002Fclarify-requirements\u002FSKILL.md",[],[66,67,68,69,70,71,72,73,74,29,75,76,77,78,79],"clarify requirements","clarify","requirements","risk","options","task","prompts","goal","scanning","context","repo","injection","execution","file",{"skillNo":81,"name":82,"nameZh":6,"description":83,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":84,"categories":85,"keywords":91,"publishedAt":33},"01M6DF79CA4BHJS66J8EZDBPWJ","openmm-md","Run FREE, open-source, Windows-native GPU molecular dynamics via the openmm-md MCP (OpenMM + CUDA on the local NVIDIA GPU). Use for protein \u002F protein-ligand MD when you want a free stack or are on Windows without a Schrodinger license. Covers the full workflow (prepare, solvate, minimize, equilibrate, production, analyze, mmgbsa) AND an honest decision guide vs Schrodinger Desmond, including how to run Desmond legitimately on Linux or a cloud GPU with a real license. This is a deliberately LESSER free alternative and does NOT replace the schrodinger MCP tools. Do NOT use it to pick a free tool when a validated FEP+, WaterMap, IFD-MD, or Prime MM-GBSA result is actually required.","https:\u002F\u002Fgithub.com\u002FPrekzursil\u002Fagent-skills-toolchain\u002Fblob\u002Fmain\u002Ftools\u002Fmcp\u002Fopenmm-md-mcp\u002Fskill\u002Fopenmm-md\u002FSKILL.md",[86],{"categoryNo":87,"categoryName":88,"taxonomyType":89,"taxonomyTypeDesc":90},"01JBSEEDCAT000000000000023","Skill与智能体开发","SCENARIO","使用场景",[92,93,94,95,96,97,98,99,100,101,102,103,104,105,29,106,107,108,109,110],"openmm md","how","decision","pick","linux","analyze","tool","stack","open","NVIDIA","gpu","license","workflow","windows","free","molecular dynamics","local","do","MCP",{"skillNo":112,"name":113,"nameZh":6,"description":114,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":115,"categories":116,"keywords":123,"publishedAt":33},"01M6DF76C1MG85VEWP8VZCBNMN","visual-review","Take screenshots of web pages and UI using Playwright for visual review and iteration","https:\u002F\u002Fgithub.com\u002FPrekzursil\u002Fagent-skills-toolchain\u002Fblob\u002Fmain\u002Ftools\u002Feditor\u002Fmetaswarm\u002Fskills\u002Fvisual-review\u002FSKILL.md",[117,120],{"categoryNo":118,"categoryName":119,"taxonomyType":89,"taxonomyTypeDesc":90},"01JBSEEDCAT000000000000003","演示文稿",{"categoryNo":121,"categoryName":122,"taxonomyType":89,"taxonomyTypeDesc":90},"01JBSEEDCAT000000000000008","测试与质量",[124,125,126,127,128,129,130,131,132,133,93,134,135,136,137],"review","visual","visual review","pages","web","false","iteration","take","check","slides","ui","screenshot","playwright","screenshots",{"skillNo":139,"name":140,"nameZh":6,"description":141,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":142,"categories":143,"keywords":144,"publishedAt":33},"01M6DF73AQVCBZKEN0YXEX9FXP","status","Diagnostic status report — shows metaswarm installation state, project setup, and potential issues","https:\u002F\u002Fgithub.com\u002FPrekzursil\u002Fagent-skills-toolchain\u002Fblob\u002Fmain\u002Ftools\u002Feditor\u002Fmetaswarm\u002Fskills\u002Fstatus\u002FSKILL.md",[],[140,145,146,147,148,149,150,24],"setup","report","project","installation","project setup","issues",{"skillNo":152,"name":153,"nameZh":6,"description":154,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":155,"categories":156,"keywords":158,"publishedAt":33},"01M6DF6X6C2XB0P6GX76P7TTYV","start","Use when starting work on any task, when the user mentions metaswarm, or when the user wants to begin tracked development work","https:\u002F\u002Fgithub.com\u002FPrekzursil\u002Fagent-skills-toolchain\u002Fblob\u002Fmain\u002Ftools\u002Feditor\u002Fmetaswarm\u002Fskills\u002Fstart\u002FSKILL.md",[157],{"categoryNo":87,"categoryName":88,"taxonomyType":89,"taxonomyTypeDesc":90},[153,159,71,160,161,162,29,163,164,165,166],"start issue","issue","ready","work on issue","user","work","work on","development",{"skillNo":168,"name":145,"nameZh":6,"description":169,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":170,"categories":171,"keywords":172,"publishedAt":33},"01M6DF6QMJ9AJKFKHHKRT005PQ","Interactive project setup — detects your project, configures metaswarm, writes project-local files","https:\u002F\u002Fgithub.com\u002FPrekzursil\u002Fagent-skills-toolchain\u002Fblob\u002Fmain\u002Ftools\u002Feditor\u002Fmetaswarm\u002Fskills\u002Fsetup\u002FSKILL.md",[],[145,147,108,173,149,174],"files","interactive"]