tooluniverse-immunotherapy-response-prediction by mims-harvard/tooluniverse
npx skills add https://github.com/mims-harvard/tooluniverse --skill tooluniverse-immunotherapy-response-prediction通过多生物标志物整合,预测患者对免疫检查点抑制剂(ICIs)的反应。将患者肿瘤特征(癌症类型 + 突变 + 生物标志物)转化为量化的 ICI 反应评分,并提供药物特异性建议、耐药风险评估和监测计划。
核心原则:
当用户询问以下问题时应用:
必需项:癌症类型 + 至少一项:突变列表 或 TMB 值 可选项:PD-L1 表达、MSI 状态、免疫浸润数据、HLA 类型、既往治疗史、拟用 ICI
有关输入格式示例、癌症类型标准化和基因符号标准化表,请参阅 INPUT_REFERENCE.md。
广告位招租
在这里展示您的产品或服务
触达数万 AI 开发者,精准高效
Input: Cancer type + Mutations/TMB + Optional biomarkers (PD-L1, MSI, etc.)
Phase 1: Input Standardization & Cancer Context
Phase 2: TMB Analysis
Phase 3: Neoantigen Analysis
Phase 4: MSI/MMR Status Assessment
Phase 5: PD-L1 Expression Analysis
Phase 6: Immune Microenvironment Profiling
Phase 7: Mutation-Based Predictors
Phase 8: Clinical Evidence & ICI Options
Phase 9: Resistance Risk Assessment
Phase 10: Multi-Biomarker Score Integration
Phase 11: Clinical Recommendations
OpenTargets_get_disease_id_description_by_name 解析癌症类型 为 EFO ID{gene, variant, type}MyGene_query_genes 解析基因 IDfda_pharmacogenomic_biomarkers(drug_name='pembrolizumab') 检查 FDA TMB-H 生物标志物UniProt_get_function_by_accession 检查突变影响iedb_search_epitopes 查询已知表位fda_pharmacogenomic_biomarkers(biomarker='Microsatellite Instability') 检查 FDA MSI-H 批准情况HPA_get_cancer_prognostics_by_gene(gene_name='CD274') 获取基线表达enrichr_gene_enrichment_analysis 运行免疫通路富集分析FDA_get_indications_by_drug_name 查询 ICI 药物的 FDA 适应症clinical_trials_search 或 search_clinical_trials 搜索临床试验OpenTargets_get_drug_mechanisms_of_action_by_chemblId 获取药物作用机制有关 ICI 药物概况和 ChEMBL ID,请参阅 SCORING_TABLES.md。
civic_search_evidence_items 在 CIViC 中检查耐药证据TOTAL SCORE = TMB_score + MSI_score + PDL1_score + Neoantigen_score + Mutation_bonus + Resistance_penalty
TMB_score: 5-30 points MSI_score: 5-25 points
PDL1_score: 5-20 points Neoantigen_score: 5-15 points
Mutation_bonus: 0-10 points Resistance_penalty: -20 to 0 points
Floor: 0, Cap: 100
反应可能性等级:
置信度:高(具备全部 4 种生物标志物)、中-高(具备 3/4)、中(具备 2/4)、低(具备 1 种)、极低(仅知癌症类型)
保存为 immunotherapy_response_prediction_{cancer_type}.md。完整的报告结构请参阅 REPORT_TEMPLATE.md。
在调用任何工具之前,请验证参数。有关已验证的工具参数表,请参阅 TOOLS_REFERENCE.md。
关键提醒:
MyGene_query_genes:使用 query(不是 q)EnsemblVEP_annotate_rsid:使用 variant_id(不是 rsid)drugbank_* 工具:所有 4 个参数都是必需的(query、case_sensitive、exact_match、limit)cBioPortal_get_mutations:gene_list 是字符串,不是数组ensembl_lookup_gene:需要 species='homo_sapiens'| 等级 | 描述 | 来源示例 |
|---|---|---|
| T1 | FDA 批准的生物标志物/适应症 | FDA 标签、NCCN 指南 |
| T2 | 2-3 期临床试验证据 | 已发表的试验数据、PubMed |
| T3 | 临床前/计算证据 | 通路分析、体外数据 |
| T4 | 专家意见/病例报告 | 病例系列、综述 |
每周安装次数
123
代码库
GitHub 星标数
1.2K
首次出现
Feb 19, 2026
安全审计
安装于
codex120
gemini-cli119
opencode119
github-copilot118
cursor116
amp115
Predict patient response to immune checkpoint inhibitors (ICIs) using multi-biomarker integration. Transforms a patient tumor profile (cancer type + mutations + biomarkers) into a quantitative ICI Response Score with drug-specific recommendations, resistance risk assessment, and monitoring plan.
KEY PRINCIPLES :
Apply when user asks:
Required : Cancer type + at least one of: mutation list OR TMB value Optional : PD-L1 expression, MSI status, immune infiltration data, HLA type, prior treatments, intended ICI
See INPUT_REFERENCE.md for input format examples, cancer type normalization, and gene symbol normalization tables.
Input: Cancer type + Mutations/TMB + Optional biomarkers (PD-L1, MSI, etc.)
Phase 1: Input Standardization & Cancer Context
Phase 2: TMB Analysis
Phase 3: Neoantigen Analysis
Phase 4: MSI/MMR Status Assessment
Phase 5: PD-L1 Expression Analysis
Phase 6: Immune Microenvironment Profiling
Phase 7: Mutation-Based Predictors
Phase 8: Clinical Evidence & ICI Options
Phase 9: Resistance Risk Assessment
Phase 10: Multi-Biomarker Score Integration
Phase 11: Clinical Recommendations
OpenTargets_get_disease_id_description_by_name{gene, variant, type}MyGene_query_genesfda_pharmacogenomic_biomarkers(drug_name='pembrolizumab')UniProt_get_function_by_accessioniedb_search_epitopesfda_pharmacogenomic_biomarkers(biomarker='Microsatellite Instability')HPA_get_cancer_prognostics_by_gene(gene_name='CD274')enrichr_gene_enrichment_analysisFDA_get_indications_by_drug_nameclinical_trials_search or search_clinical_trialsOpenTargets_get_drug_mechanisms_of_action_by_chemblIdSee SCORING_TABLES.md for ICI drug profiles and ChEMBL IDs.
civic_search_evidence_itemsTOTAL SCORE = TMB_score + MSI_score + PDL1_score + Neoantigen_score + Mutation_bonus + Resistance_penalty
TMB_score: 5-30 points MSI_score: 5-25 points
PDL1_score: 5-20 points Neoantigen_score: 5-15 points
Mutation_bonus: 0-10 points Resistance_penalty: -20 to 0 points
Floor: 0, Cap: 100
Response Likelihood Tiers :
Confidence : HIGH (all 4 biomarkers), MODERATE-HIGH (3/4), MODERATE (2/4), LOW (1), VERY LOW (cancer only)
Save as immunotherapy_response_prediction_{cancer_type}.md. See REPORT_TEMPLATE.md for the full report structure.
BEFORE calling ANY tool , verify parameters. See TOOLS_REFERENCE.md for verified tool parameters table.
Key reminders:
MyGene_query_genes: use query (NOT q)EnsemblVEP_annotate_rsid: use variant_id (NOT rsid)drugbank_* tools: ALL 4 params required (query, case_sensitive, exact_match, limit)| Tier | Description | Source Examples |
|---|---|---|
| T1 | FDA-approved biomarker/indication | FDA labels, NCCN guidelines |
| T2 | Phase 2-3 clinical trial evidence | Published trial data, PubMed |
| T3 | Preclinical/computational evidence | Pathway analysis, in vitro data |
| T4 | Expert opinion/case reports | Case series, reviews |
Weekly Installs
123
Repository
GitHub Stars
1.2K
First Seen
Feb 19, 2026
Security Audits
Gen Agent Trust HubPassSocketPassSnykWarn
Installed on
codex120
gemini-cli119
opencode119
github-copilot118
cursor116
amp115
AI 代码实施计划编写技能 | 自动化开发任务分解与 TDD 流程规划工具
49,000 周安装
Notion知识捕获工具:将对话笔记自动转为结构化Notion页面,提升团队知识管理效率
2 周安装
视觉审计工具:自动检查Quarto、Beamer、Typst文档布局问题,优化排版与间距
2 周安装
文档校对工具proofread - 自动检查语法拼写、格式一致性,生成详细质量报告
2 周安装
Google Tasks 命令行工具 - 高效管理任务的 CLI 工具,支持列表查看、添加、更新和删除
2 周安装
finish 文档自动化完成流程:Typst、Quarto、Python 编译审阅修复工具
2 周安装
Claude AI 会话状态监控工具 - context-status 检查上下文使用量、活动计划与保存状态
2 周安装
cBioPortal_get_mutations: gene_list is a STRING not arrayensembl_lookup_gene: REQUIRES species='homo_sapiens'