@bgicli/bgicli 2.1.1 → 2.2.1

package/data/skills/trial-eligibility-agent/SKILL.md

@@ -0,0 +1,54 @@
+<!--
+# COPYRIGHT NOTICE
+# This file is part of the "Universal Biomedical Skills" project.
+# Copyright (c) 2026 MD BABU MIA, PhD <md.babu.mia@mssm.edu>
+# All Rights Reserved.
+#
+# This code is proprietary and confidential.
+# Unauthorized copying of this file, via any medium is strictly prohibited.
+#
+# Provenance: Authenticated by MD BABU MIA
+
+-->
+
+---
+name: trial-eligibility-agent
+description: Parse trial protocols and patient data to produce criterion-level MET/NOT/UNKNOWN determinations with evidence and gaps for clinical trial screening tasks.
+allowed-tools:
+  - read_file
+  - run_shell_command
+---
+
+## At-a-Glance
+- **description (10-20 chars):** Trial triage hub
+- **keywords:** eligibility, ClinicalTrials, FHIR, evidence, gaps
+- **measurable_outcome:** Produce a MET/NOT/UNKNOWN matrix with supporting citations for ≥90% of inclusion/exclusion criteria within 5 minutes per trial request.
+
+## Inputs
+- `trial_id` (NCT or sponsor ID) plus protocol text if not public.
+- `patient_summary` narrative and optional `patient_structured` FHIR bundle.
+- Declare data sources used (notes, labs, imaging, meds) to show provenance.
+
+## Outputs
+1. Structured table (JSON recommended) listing each criterion id/text with status, evidence snippet, and confidence.
+2. Overall recommendation (`potentially_eligible`, `not_eligible`, `needs_more_information`).
+3. Data gap checklist covering missing labs/imaging/biomarkers.
+
+## Workflow
+1. **Acquire protocol:** Pull eligibility text from ClinicalTrials.gov or sponsor PDF.
+2. **Normalize criteria:** Break into atomic checks with AND/OR logic and thresholds.
+3. **Extract patient facts:** Map narrative + FHIR data into canonical features (age, labs, ECOG, biomarkers).
+4. **Evaluate:** Assign MET/NOT/UNKNOWN with cited evidence for each criterion, flag missing context explicitly.
+5. **Summarize:** Present recommendation and highlight gating unknowns plus next-best actions.
+
+## Guardrails
+- Never claim enrollment decisions; mark outputs as advisory.
+- Cite direct patient evidence for every MET/NOT call; default to UNKNOWN rather than guessing.
+- Respect PHI handling expectations—avoid storing raw notes outside secure paths.
+
+## Tooling & References
+- Use `README.md` for API snippets (FHIR parsing, JSON schema) and dependency versions.
+- Pair with `Clinical/Trial_Matching/TrialGPT` when retrieval/ranking is also needed.
+
+
+<!-- AUTHOR_SIGNATURE: 9a7f3c2e-MD-BABU-MIA-2026-MSSM-SECURE -->

package/data/skills/trialgpt-matching/SKILL.md

@@ -0,0 +1,66 @@
+<!--
+# COPYRIGHT NOTICE
+# This file is part of the "Universal Biomedical Skills" project.
+# Copyright (c) 2026 MD BABU MIA, PhD <md.babu.mia@mssm.edu>
+# All Rights Reserved.
+#
+# This code is proprietary and confidential.
+# Unauthorized copying of this file, via any medium is strictly prohibited.
+#
+# Provenance: Authenticated by MD BABU MIA
+
+-->
+
+---
+name: trialgpt-matching
+description: Trial shortlist
+keywords:
+  - retrieval
+  - ranking
+  - ClinicalTrials
+  - patient-profile
+measurable_outcome: Produce ≥5 ranked trials (when available) with rationale + missing-data notes within 3 minutes of receiving a patient query.
+license: MIT
+metadata:
+  author: TrialGPT Team
+  version: "1.0.0"
+compatibility:
+  - system: Python 3.9+
+allowed-tools:
+  - run_shell_command
+  - read_file
+---
+
+# TrialGPT Matching
+
+Run the locally checked-out TrialGPT pipeline to retrieve, rank, and explain candidate trials for a patient before deeper eligibility review.
+
+## Inputs
+- Patient summary (structured JSON or free text) with condition keywords.
+- Optional filters: geography, phase, intervention, biomarker.
+- Up-to-date ClinicalTrials.gov dump or API access.
+
+## Outputs
+- Ranked trial table with NCT ID, title, score, and short justification.
+- Parsed inclusion/exclusion text ready for downstream eligibility agents.
+- Missing data checklist (e.g., "ECOG not provided").
+
+## Workflow
+1. **Setup:** `cd repo && pip install -r requirements.txt` (or reuse env).
+2. **Trial retrieval:** Run TrialGPT retriever to pull candidate trials for the indication.
+3. **Criteria parsing:** Convert eligibility blocks to structured criteria JSON.
+4. **Patient profiling:** Summarize patient facts (labs, prior therapies, biomarkers).
+5. **Ranking:** Execute TrialGPT ranking script to score each trial and emit explanations.
+6. **Handoff:** Export ranked list + structured criteria for `trial-eligibility-agent`.
+
+## Guardrails
+- Refresh ClinicalTrials.gov metadata regularly to avoid stale trials.
+- Label scores as AI-generated suggestions pending clinician validation.
+- Retain prompt/config metadata for audit trails.
+
+## References
+- Detailed usage instructions and repo layout live in `README.md`.
+- Coordinate with `Skills/Clinical/Trial_Eligibility_Agent` for criterion-level review.
+
+
+<!-- AUTHOR_SIGNATURE: 9a7f3c2e-MD-BABU-MIA-2026-MSSM-SECURE -->

package/data/skills/tumor-clonal-evolution-agent/SKILL.md

@@ -0,0 +1,134 @@
+<!--
+# COPYRIGHT NOTICE
+# This file is part of the "Universal Biomedical Skills" project.
+# Copyright (c) 2026 MD BABU MIA, PhD <md.babu.mia@mssm.edu>
+# All Rights Reserved.
+#
+# This code is proprietary and confidential.
+# Unauthorized copying of this file, via any medium is strictly prohibited.
+#
+# Provenance: Authenticated by MD BABU MIA
+
+-->
+
+---
+name: 'tumor-clonal-evolution-agent'
+description: 'AI-powered analysis of tumor clonal architecture, subclonal dynamics, and evolutionary trajectories from multi-region sequencing and longitudinal liquid biopsy data.'
+measurable_outcome: Execute skill workflow successfully with valid output within 15 minutes.
+allowed-tools:
+  - read_file
+  - run_shell_command
+---
+
+
+# Tumor Clonal Evolution Agent
+
+The **Tumor Clonal Evolution Agent** analyzes intratumoral heterogeneity (ITH), reconstructs tumor phylogenies, and tracks clonal dynamics over time. It integrates multi-region sequencing data, longitudinal liquid biopsies, and mathematical modeling to predict treatment response and resistance emergence.
+
+## When to Use This Skill
+
+* When analyzing multi-region tumor sequencing to map spatial heterogeneity.
+* To reconstruct tumor phylogenetic trees and identify ancestral mutations.
+* For tracking clonal evolution through serial liquid biopsy samples.
+* To predict time to treatment failure using evolutionary modeling.
+* When identifying resistance-conferring subclones before clinical progression.
+
+## Core Capabilities
+
+1. **Clonal Deconvolution**: Identifies tumor subpopulations and estimates their cellular fractions using variant allele frequencies (VAF) from bulk sequencing.
+
+2. **Phylogenetic Reconstruction**: Builds tumor evolutionary trees showing relationships between subclones and their mutational acquisition order.
+
+3. **Longitudinal Tracking**: Monitors subclone dynamics over time using ctDNA variant frequencies from serial blood draws.
+
+4. **Resistance Prediction**: Applies Bayesian evolutionary frameworks to forecast emergence of resistant clones and time to progression.
+
+5. **Spatial ITH Mapping**: Integrates multi-region data to visualize spatial distribution of subclones across tumor sites.
+
+6. **Fitness Estimation**: Calculates subclone fitness parameters to identify aggressive populations driving tumor progression.
+
+## Workflow
+
+1. **Input**: Multi-region or longitudinal mutation data (VCF/MAF), tumor purity estimates, copy number profiles.
+
+2. **Clustering**: Cluster mutations into subclones using PyClone, SciClone, or MOBSTER.
+
+3. **Phylogeny**: Reconstruct evolutionary trees using CITUP, PhyloWGS, or CALDER.
+
+4. **Modeling**: Apply mathematical models (Lotka-Volterra, birth-death) to estimate dynamics.
+
+5. **Prediction**: Forecast treatment response and resistance timeline.
+
+6. **Output**: Phylogenetic trees, subclone trajectories, resistance predictions, actionable insights.
+
+## Example Usage
+
+**User**: "Analyze the clonal evolution from these 6 longitudinal ctDNA samples and predict time to progression."
+
+**Agent Action**:
+```bash
+python3 Skills/Oncology/Tumor_Clonal_Evolution_Agent/clonal_evolution.py \
+    --input longitudinal_ctdna_variants.maf \
+    --timepoints 0,4,8,12,16,20 \
+    --tumor_burden cea_values.csv \
+    --method bayesian_evolution \
+    --predict_ttp true \
+    --output evolution_analysis/
+```
+
+## Key Methods and Algorithms
+
+| Tool/Method | Application | Reference |
+|-------------|-------------|-----------|
+| PyClone-VI | Bayesian clustering of mutations | Nature Methods 2014 |
+| MOBSTER | Subclonal deconvolution with selection | Nature Genetics 2020 |
+| PhyloWGS | Phylogenetic tree reconstruction | Genome Biology 2015 |
+| CALDER | Copy-number aware phylogeny | Nature Methods 2019 |
+| CHESS | Cancer heterogeneity from single samples | Cell Systems 2019 |
+
+## Mathematical Framework
+
+The agent applies evolutionary dynamics models:
+
+**Lotka-Volterra Competition**:
+```
+dNi/dt = ri * Ni * (1 - sum(aij * Nj) / Ki)
+```
+
+Where:
+- Ni = population of subclone i
+- ri = growth rate (fitness)
+- aij = competition coefficient
+- Ki = carrying capacity
+
+**VAF Dynamics Modeling**:
+- Serial ctDNA VAF measurements enable real-time fitness estimation
+- Bayesian inference updates subclone parameters with each sample
+- Monte Carlo simulations generate prediction intervals
+
+## Prerequisites
+
+* Python 3.10+
+* PyClone-VI, PhyloWGS, or MOBSTER
+* Copy number calling tools (ASCAT, Sequenza)
+* Statistical modeling (PyMC, Stan)
+
+## Related Skills
+
+* ctDNA_Analysis - For cfDNA variant calling
+* Liquid_Biopsy_Analysis - For blood-based biomarker detection
+* Variant_Interpretation - For mutation annotation
+
+## Clinical Applications
+
+1. **Treatment Selection**: Identify dominant subclones to target
+2. **Resistance Monitoring**: Detect emerging resistant populations early
+3. **Prognosis**: Predict time to treatment failure
+4. **Combination Therapy**: Design strategies targeting multiple subclones
+
+## Author
+
+AI Group - Biomedical AI Platform
+
+
+<!-- AUTHOR_SIGNATURE: 9a7f3c2e-MD-BABU-MIA-2026-MSSM-SECURE -->

package/data/skills/tumor-heterogeneity-agent/SKILL.md

@@ -0,0 +1,216 @@
+<!--
+# COPYRIGHT NOTICE
+# This file is part of the "Universal Biomedical Skills" project.
+# Copyright (c) 2026 MD BABU MIA, PhD <md.babu.mia@mssm.edu>
+# All Rights Reserved.
+#
+# This code is proprietary and confidential.
+# Unauthorized copying of this file, via any medium is strictly prohibited.
+#
+# Provenance: Authenticated by MD BABU MIA
+
+-->
+
+---
+name: 'tumor-heterogeneity-agent'
+description: 'AI-powered intratumor heterogeneity analysis for clonal architecture reconstruction, subclonal evolution tracking, and therapy resistance prediction using multi-region and longitudinal sequencing.'
+measurable_outcome: Execute skill workflow successfully with valid output within 15 minutes.
+allowed-tools:
+  - read_file
+  - run_shell_command
+---
+
+
+# Tumor Heterogeneity Agent
+
+The **Tumor Heterogeneity Agent** provides comprehensive analysis of intratumor heterogeneity (ITH) for understanding clonal architecture, tracking subclonal evolution, and predicting therapy resistance. It integrates multi-region sequencing, single-cell data, and longitudinal samples to reconstruct tumor phylogenies and identify actionable subclones.
+
+## When to Use This Skill
+
+* When analyzing multi-region tumor sequencing for clonal architecture.
+* For tracking clonal evolution under treatment pressure.
+* To predict resistance emergence from subclonal populations.
+* When assessing tumor heterogeneity impact on treatment response.
+* For integrating single-cell and bulk sequencing for ITH analysis.
+
+## Core Capabilities
+
+1. **Clonal Deconvolution**: Infer clonal populations and their frequencies.
+
+2. **Phylogeny Reconstruction**: Build tumor evolutionary trees from variants.
+
+3. **Subclonal Tracking**: Monitor subclone dynamics over time.
+
+4. **Resistance Prediction**: Identify pre-existing resistant subclones.
+
+5. **Multi-Region Integration**: Combine spatial heterogeneity data.
+
+6. **Single-Cell ITH**: Integrate scDNA-seq for ground-truth clones.
+
+## Heterogeneity Metrics
+
+| Metric | Definition | Clinical Relevance |
+|--------|------------|-------------------|
+| MATH Score | Mutant-allele tumor heterogeneity | ITH quantification |
+| Shannon Index | Clonal diversity | Evolutionary potential |
+| Clone Count | Number of distinct clones | Complexity |
+| Truncal Fraction | % truncal mutations | Targetability |
+| ITH Score | Composite heterogeneity | Prognosis |
+
+## Workflow
+
+1. **Input**: Multi-region/longitudinal WES/WGS, copy number, tumor purity.
+
+2. **Preprocessing**: Variant calling, CNV calling, purity estimation.
+
+3. **CCF Estimation**: Calculate cancer cell fraction for each mutation.
+
+4. **Clustering**: Group mutations into clonal populations.
+
+5. **Phylogeny**: Reconstruct evolutionary tree.
+
+6. **Temporal Analysis**: Track clone dynamics over time.
+
+7. **Output**: Clone structures, phylogenies, heterogeneity metrics.
+
+## Example Usage
+
+**User**: "Analyze the clonal architecture of this multi-region lung tumor sequencing to understand heterogeneity and identify resistant subclones."
+
+**Agent Action**:
+```bash
+python3 Skills/Oncology/Tumor_Heterogeneity_Agent/ith_analysis.py \
+    --multi_region_vcfs region1.vcf,region2.vcf,region3.vcf \
+    --cnv_segments cnv_calls.seg \
+    --purity 0.7,0.65,0.72 \
+    --sample_names Primary,Met1,Met2 \
+    --method pyclone-vi \
+    --phylogeny_method citup \
+    --output ith_analysis/
+```
+
+## Deconvolution Methods
+
+| Method | Approach | Best For |
+|--------|----------|----------|
+| PyClone-VI | Variational inference | Large datasets |
+| SciClone | Kernel density | High purity |
+| EXPANDS | Probabilistic | Multi-region |
+| Canopy | EM algorithm | CNV integration |
+| Clonevol | Phylogeny-aware | Longitudinal |
+| CITUP | Integer programming | Tree optimization |
+
+## Input Requirements
+
+| Input | Format | Required |
+|-------|--------|----------|
+| Somatic Variants | VCF with depth | Yes |
+| Copy Number | SEG file | Yes |
+| Tumor Purity | Float (0-1) | Yes |
+| Sample Metadata | TSV | Yes |
+| Normal BAM | BAM | Recommended |
+
+## Output Components
+
+| Output | Description | Format |
+|--------|-------------|--------|
+| Clone Assignments | Mutation-to-clone mapping | .csv |
+| Clone Frequencies | Per-sample clone fractions | .csv |
+| Phylogenetic Tree | Newick and visualization | .nwk, .pdf |
+| ITH Metrics | Heterogeneity scores | .json |
+| Subclone Variants | Clone-specific mutations | .vcf |
+| Evolution Plot | Clone dynamics over time | .png |
+| Actionable Subclones | Druggable clone mutations | .csv |
+
+## Clonal Classification
+
+| Clone Type | Definition | Implications |
+|------------|------------|--------------|
+| Truncal | Present in all samples | Ideal targets |
+| Branch | Present in subset | Regional targets |
+| Private | Single sample only | Local significance |
+| Resistant | Expand under therapy | Resistance mechanism |
+
+## AI/ML Components
+
+**Clone Inference**:
+- Variational autoencoders for CCF estimation
+- Dirichlet process mixture models
+- Graph neural networks for phylogeny
+
+**Resistance Prediction**:
+- Time-series models for clone trajectories
+- Classification of resistant signatures
+- Drug-clone interaction prediction
+
+**Multi-Region Integration**:
+- Multi-task learning across regions
+- Spatial models for regional patterns
+- Transfer learning across cancers
+
+## Clinical Applications
+
+| Application | ITH Insight | Clinical Action |
+|-------------|-------------|-----------------|
+| Treatment Selection | Truncal vs branch targets | Prioritize truncal targets |
+| Resistance Monitoring | Pre-existing resistant clones | Early combination therapy |
+| Prognosis | ITH score | Risk stratification |
+| Biomarker Development | Clonal biomarkers | Robust biomarker selection |
+
+## Cancer-Specific Patterns
+
+| Cancer Type | Typical ITH | Key Drivers |
+|-------------|-------------|-------------|
+| Lung (NSCLC) | High | EGFR, KRAS subclonal |
+| Breast | Moderate-High | PIK3CA, ESR1 evolution |
+| Colorectal | Moderate | KRAS, BRAF clonal |
+| Renal | Very High | VHL truncal, diverse branches |
+| Melanoma | High | BRAF/NRAS truncal |
+
+## Prerequisites
+
+* Python 3.10+
+* PyClone-VI, SciClone
+* CITUP, Clonevol
+* CNVkit/FACETS for CNV
+* R with clonal evolution packages
+
+## Related Skills
+
+* ctDNA_Dynamics_MRD_Agent - Liquid biopsy tracking
+* Single_Cell_CNV_Agent - scDNA-seq analysis
+* HRD_Analysis_Agent - Genomic instability
+* Pan_Cancer_MultiOmics_Agent - Multi-omic integration
+
+## Phylogeny Visualization
+
+| View Type | Shows | Best For |
+|-----------|-------|----------|
+| Fish Plot | Clone dynamics over time | Longitudinal |
+| Tree Diagram | Branching evolution | Multi-region |
+| Muller Plot | Population dynamics | Treatment response |
+| Clone Map | Spatial distribution | Multi-region spatial |
+
+## Special Considerations
+
+1. **Sampling Bias**: Multi-region captures more heterogeneity
+2. **Purity Effects**: Low purity reduces clone resolution
+3. **CNV Complexity**: High CNV burden complicates CCF
+4. **Single-Cell Validation**: Ground truth from scDNA-seq
+5. **Temporal Resolution**: Frequent sampling improves tracking
+
+## Resistance Mechanisms
+
+| Mechanism | Detection | Intervention |
+|-----------|-----------|--------------|
+| Pre-existing resistant clone | Subclonal at baseline | Combination therapy |
+| Acquired resistance | New clone emerges | Switch therapy |
+| Phenotypic plasticity | Expression change | Monitor phenotype |
+| Microenvironment | TME evolution | Immunotherapy |
+
+## Author
+
+AI Group - Biomedical AI Platform
+
+
+<!-- AUTHOR_SIGNATURE: 9a7f3c2e-MD-BABU-MIA-2026-MSSM-SECURE -->

package/data/skills/tumor-mutational-burden-agent/SKILL.md

@@ -0,0 +1,188 @@
+<!--
+# COPYRIGHT NOTICE
+# This file is part of the "Universal Biomedical Skills" project.
+# Copyright (c) 2026 MD BABU MIA, PhD <md.babu.mia@mssm.edu>
+# All Rights Reserved.
+#
+# This code is proprietary and confidential.
+# Unauthorized copying of this file, via any medium is strictly prohibited.
+#
+# Provenance: Authenticated by MD BABU MIA
+
+-->
+
+---
+name: 'tumor-mutational-burden-agent'
+description: 'Calculates and harmonizes Tumor Mutational Burden (TMB) across platforms to predict immunotherapy response.'
+keywords:
+  - tmb
+  - immunotherapy
+  - biomarker
+  - harmonization
+  - oncology
+measurable_outcome: 'Harmonizes TMB scores across 5+ assay platforms with <5% variance from WES gold standard.'
+allowed-tools:
+  - read_file
+  - run_shell_command
+---
+
+
+# Tumor Mutational Burden Agent
+
+The **Tumor Mutational Burden Agent** provides comprehensive TMB analysis for immunotherapy response prediction. It harmonizes TMB calculation across different assays, integrates with other biomarkers (PD-L1, MSI), and provides evidence-based therapy recommendations.
+
+## When to Use This Skill
+
+* When calculating TMB from panel sequencing, WES, or WGS data.
+* To harmonize TMB values across different assay platforms.
+* For predicting immunotherapy response using TMB and integrated biomarkers.
+* When determining TMB-High status for pembrolizumab eligibility.
+* To analyze TMB in context of tumor type-specific distributions.
+
+## Core Capabilities
+
+1. **TMB Calculation**: Compute TMB from different sequencing platforms with appropriate normalization.
+
+2. **Platform Harmonization**: Standardize TMB across FoundationOne, MSK-IMPACT, WES, and other assays.
+
+3. **TMB-High Classification**: Apply FDA-approved and tumor-specific thresholds.
+
+4. **Biomarker Integration**: Combine TMB with PD-L1, MSI, and gene signatures.
+
+5. **Response Prediction**: ML models predicting ICI response from TMB-inclusive features.
+
+6. **Tumor-Specific Context**: Interpret TMB relative to cancer type distributions.
+
+## TMB Calculation Methods
+
+| Platform | Coverage | TMB Formula | Normalization |
+|----------|----------|-------------|---------------|
+| WES | 30-50 Mb | Nonsynonymous/coding Mb | Per exome size |
+| FoundationOne | 1.1 Mb | Syn + nonsyn/panel Mb | FDA validated |
+| MSK-IMPACT | 1.0-1.2 Mb | Nonsyn + splice/panel Mb | Panel-specific |
+| TSO500 | 1.94 Mb | Coding mutations/Mb | Illumina validated |
+| WGS | 3 Gb | Various metrics | Genome-wide |
+
+## TMB Thresholds
+
+| Context | Threshold | Evidence |
+|---------|-----------|----------|
+| FDA (pan-tumor) | ≥10 mut/Mb | KEYNOTE-158 |
+| Melanoma | ≥10 mut/Mb | Practice standard |
+| NSCLC | ≥10 mut/Mb | Multiple trials |
+| SCLC | ≥10 mut/Mb | Variable benefit |
+| Colorectal (MSS) | Limited utility | MSI more predictive |
+| Urothelial | ≥10 mut/Mb | IMvigor trials |
+
+## Workflow
+
+1. **Input**: VCF/MAF file with somatic mutations, assay details, tumor type.
+
+2. **Filtering**: Remove germline, artifacts, known drivers (optional).
+
+3. **Calculation**: Count mutations and normalize to coverage.
+
+4. **Harmonization**: Convert to WES-equivalent TMB if needed.
+
+5. **Classification**: Assign TMB-High/Low based on thresholds.
+
+6. **Integration**: Combine with PD-L1, MSI for composite score.
+
+7. **Output**: TMB value, classification, response prediction, recommendations.
+
+## Example Usage
+
+**User**: "Calculate TMB from this panel sequencing data and predict immunotherapy response."
+
+**Agent Action**:
+```bash
+python3 Skills/Oncology/Tumor_Mutational_Burden_Agent/tmb_analyzer.py \
+    --mutations tumor_somatic.maf \
+    --panel foundation_one \
+    --tumor_type nsclc \
+    --pdl1_tps 50 \
+    --msi_status stable \
+    --harmonize_to wes \
+    --output tmb_report.json
+```
+
+## Platform Harmonization
+
+Different panels yield different TMB values for the same tumor:
+
+```
+TMB_WES = a * TMB_panel + b
+
+Conversion factors (example):
+- FoundationOne CDx: TMB_WES ≈ 1.0 × TMB_F1
+- MSK-IMPACT: TMB_WES ≈ 1.1 × TMB_IMPACT
+- TSO500: TMB_WES ≈ 0.9 × TMB_TSO
+```
+
+**Harmonization Considerations**:
+- Panel size affects precision
+- Gene content affects which mutations counted
+- Algorithmic differences in filtering
+
+## Integrated Biomarker Analysis
+
+**TMB + PD-L1 + MSI Integration**:
+
+| TMB | PD-L1 | MSI | ICI Benefit |
+|-----|-------|-----|-------------|
+| High | High | MSI-H | Very high |
+| High | Low | MSS | Moderate-high |
+| Low | High | MSS | Moderate |
+| Low | Low | MSS | Limited |
+| Any | Any | MSI-H | High (pembrolizumab) |
+
+## Cancer Type TMB Distributions
+
+| Cancer Type | Median TMB | TMB-High % |
+|-------------|------------|------------|
+| Melanoma | 13.5 | 45% |
+| NSCLC | 7.2 | 25% |
+| SCLC | 9.8 | 35% |
+| Bladder | 6.5 | 20% |
+| Colorectal | 4.0 | 5% (MSS) |
+| Breast | 2.5 | 5% |
+| Prostate | 2.0 | 3% |
+
+## AI/ML Enhancement
+
+**Response Prediction Model**:
+- Features: TMB, PD-L1, MSI, gene expression signatures
+- Additional: Clonal vs subclonal TMB, driver mutations
+- Performance: AUC 0.70-0.80 across tumor types
+
+**TMB Components Analysis**:
+- Clonal TMB: Mutations in all cells
+- Subclonal TMB: Mutations in subpopulations
+- Clonal TMB more predictive of response
+
+## Prerequisites
+
+* Python 3.10+
+* Variant annotation tools
+* Panel BED files for coverage
+* Reference mutation databases
+
+## Related Skills
+
+* Variant_Annotation - For mutation calling
+* Liquid_Biopsy_Analytics_Agent - For blood-based TMB
+* Immune_Checkpoint_Combination_Agent - For ICI selection
+
+## Clinical Decision Support
+
+1. **TMB-H Pembrolizumab**: FDA-approved pan-tumor indication
+2. **TMB + PD-L1**: Combined scoring for NSCLC
+3. **TMB Monitoring**: Track under immunotherapy
+4. **TMB Heterogeneity**: Consider multiple samples
+
+## Author
+
+AI Group - Biomedical AI Platform
+
+
+<!-- AUTHOR_SIGNATURE: 9a7f3c2e-MD-BABU-MIA-2026-MSSM-SECURE -->