@bgicli/bgicli 2.1.1 → 2.2.0

package/data/skills/mpn-research-assistant/SKILL.md

@@ -0,0 +1,197 @@
+<!--
+# 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: mpn-research-assistant
+description: "Myeloproliferative neoplasm (MPN) research expertise including JAK2/CALR/MPL mutations, myelofibrosis, polycythemia vera, essential thrombocythemia. Use for MPN literature search, driver mutation analysis, PPM1D pathway analysis, fibrosis markers, megakaryocyte biology, clinical trial data interpretation, and translational research."
+license: Proprietary
+---
+
+# MPN Research Assistant
+
+## Disease Classification
+
+### WHO 2022 Classification
+- **Polycythemia Vera (PV)**: JAK2V617F (95%), JAK2 exon 12 (3%)
+- **Essential Thrombocythemia (ET)**: JAK2V617F (55%), CALR (25%), MPL (5%)
+- **Primary Myelofibrosis (PMF)**: JAK2V617F (55%), CALR (25%), MPL (8%)
+- **Pre-PMF**: Early fibrotic stage, better prognosis
+- **Overt PMF**: Grade 2-3 fibrosis, splenomegaly
+
+### Driver Mutations
+
+| Mutation | Gene Location | Mechanism | VAF Significance |
+|----------|--------------|-----------|------------------|
+| JAK2V617F | 9p24.1 | Constitutive JAK-STAT activation | >50% → poor prognosis |
+| CALR (type 1) | 19p13.2 | 52bp deletion, MPL activation | Better prognosis |
+| CALR (type 2) | 19p13.2 | 5bp insertion, MPL activation | Intermediate |
+| MPL W515L/K | 1p34.2 | TPO-independent signaling | Thrombocytosis |
+
+### High Molecular Risk (HMR) Mutations
+- ASXL1, EZH2, SRSF2, IDH1/2, U2AF1
+- ≥2 HMR mutations = very high risk
+
+## PPM1D Pathway Analysis
+
+### Expression Patterns
+- 83.9-fold overexpression vs normal donors (p=0.0002)
+- JAK2V617F+ > CALR+ expression (43.4x vs 13.4x, p=0.01)
+- Mutation frequency: 1.9% (8th most common in MPNs)
+
+### Therapeutic Targets
+```python
+ppm1d_targets = {
+    'PPM1D inhibitors': ['GSK2830371', 'SL-176'],
+    'MDM2 inhibitors': ['navtemadlin (KRT-232)', 'idasanutlin'],
+    'Combination': ['PPM1D + MDM2 (synergistic)'],
+}
+
+# p53 pathway restoration
+mechanism = """
+PPM1D inhibition → ↑p53 phosphorylation → 
+↑p53 stabilization → ↑DNA damage response → 
+↑apoptosis in mutant clones
+"""
+```
+
+### Clinical Trial Data (BOREAS)
+- Navtemadlin: 15% SVR35, 24% TSS50
+- CD34+ reduction: 68-76% at 24-36 weeks
+- VAF reduction: 21% achieved ≥50% decrease
+- Fibrosis improvement: 45% by one grade
+
+## Megakaryocyte Subtypes in MPNs
+
+```python
+mk_subtypes = {
+    'Endomitotic MKs': {
+        'markers': ['ITGA2B', 'GP1BA', 'PF4', 'TUBB1'],
+        'function': 'Polyploidization',
+        'mpn_change': 'Dysregulated endomitosis'
+    },
+    'Platelet-Generating MKs': {
+        'markers': ['VWF', 'F2R', 'GP9', 'SELP'],
+        'function': 'Proplatelet formation',
+        'mpn_change': 'Abnormal platelet production'
+    },
+    'HSC Niche-Supporting MKs': {
+        'markers': ['THPO', 'IGF1', 'CXCL12', 'ANGPT1'],
+        'function': 'HSC maintenance',
+        'mpn_change': 'Disrupted niche signaling'
+    },
+    'Inflammatory MKs': {
+        'markers': ['S100A8', 'S100A9', 'CHI3L1', 'CXCL8'],
+        'function': 'Inflammation',
+        'mpn_change': 'Expanded in MF'
+    }
+}
+```
+
+## Fibrosis Markers
+
+### Psaila 2020 Fibrosis Gene Signature
+```python
+fibrosis_genes = [
+    'TGFB1', 'IL12A', 'IL1B', 'RAB37', 'TIMP1', 'APIP', 'PF4V1', 'VEGFA',
+    'FBLN2', 'SFRP1', 'COL6A2', 'COL4A2', 'COL5A1', 'PDGFRB', 'LOXL2', 'RUNX2'
+]
+
+# ECM remodeling
+ecm_markers = ['COL1A1', 'COL3A1', 'FN1', 'LAMA1', 'LAMB1']
+
+# Profibrotic cytokines
+cytokines = ['TGFB1', 'PDGF', 'VEGFA', 'IL1B', 'IL6', 'TNF']
+```
+
+## Prognostic Scoring Systems
+
+### MIPSS70+ v2.0 (Myelofibrosis)
+
+| Variable | Points |
+|----------|--------|
+| Hemoglobin <10 g/dL | 2 |
+| Blasts ≥2% | 1 |
+| Constitutional symptoms | 2 |
+| Absence of CALR type-1 | 2 |
+| HMR mutations | 2 each |
+| Unfavorable karyotype | 3 |
+
+### Risk Categories
+- Very Low: 0-1 points (10yr OS: 92%)
+- Low: 2-4 points
+- Intermediate: 5-8 points
+- High: 9-11 points
+- Very High: ≥12 points
+
+## Data Integration Template
+
+```python
+def create_mpn_patient_matrix(clinical_df, mutations_df, 
+                               cytokines_df, flow_df, degs_df):
+    """Integrate multi-modal MPN patient data."""
+    
+    # Merge clinical
+    matrix = clinical_df.copy()
+    
+    # Add mutation status
+    driver_muts = ['JAK2', 'CALR', 'MPL']
+    hmr_muts = ['ASXL1', 'EZH2', 'SRSF2', 'IDH1', 'IDH2']
+    
+    for mut in driver_muts + hmr_muts:
+        if mut in mutations_df.columns:
+            matrix[f'{mut}_status'] = mutations_df[mut]
+    
+    # Calculate HMR count
+    matrix['HMR_count'] = matrix[[f'{m}_status' for m in hmr_muts 
+                                   if f'{m}_status' in matrix.columns]].sum(axis=1)
+    
+    # Add cytokine data
+    for cyto in ['TGFB1', 'IL6', 'IL8']:
+        if cyto in cytokines_df.columns:
+            matrix[f'{cyto}_level'] = cytokines_df[cyto]
+    
+    # Add flow cytometry
+    matrix['CD34_percent'] = flow_df['CD34_positive_percent']
+    
+    return matrix
+```
+
+## Key References
+
+```python
+key_papers = {
+    'Williams_2022': 'Blood: Phylogenetic reconstruction of MPN evolution',
+    'Psaila_2020': 'Nature Medicine: Single-cell profiling of MF megakaryocytes',
+    'Mascarenhas_2022': 'Blood Advances: Idasanutlin in PV',
+    'BOREAS_2024': 'Phase III navtemadlin in MF',
+    'Marcellino_iPSC': 'PPM1D iPSC modeling in MPNs',
+    'Kanagal-Shamanna': 'Mod Pathol: i(17q) in MDS/MPN'
+}
+```
+
+## PubMed Search Templates
+
+```python
+mpn_search_queries = {
+    'ppm1d_mpn': '"PPM1D"[Title/Abstract] AND ("myeloproliferative"[Title/Abstract] OR "myelofibrosis"[Title/Abstract])',
+    'single_cell_mpn': '"single-cell"[Title/Abstract] AND "myeloproliferative neoplasm"[Title/Abstract]',
+    'jak2_calr': '(JAK2V617F OR "CALR mutation") AND myeloproliferative',
+    'fibrosis_mk': 'megakaryocyte[Title/Abstract] AND fibrosis[Title/Abstract] AND myelofibrosis'
+}
+```
+
+See `references/mpn_clinical_trials.md` for ongoing trials.
+See `references/mpn_mutations_database.md` for complete mutation catalog.
+
+
+<!-- AUTHOR_SIGNATURE: 9a7f3c2e-MD-BABU-MIA-2026-MSSM-SECURE -->

package/data/skills/mrd-edge-detection-agent/SKILL.md

@@ -0,0 +1,213 @@
+<!--
+# 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: 'mrd-edge-detection-agent'
+description: 'Ultra-sensitive AI-powered molecular residual disease detection using MRD-EDGE deep learning for sub-0.001% VAF ctDNA detection and early relapse prediction.'
+measurable_outcome: Execute skill workflow successfully with valid output within 15 minutes.
+allowed-tools:
+  - read_file
+  - run_shell_command
+---
+
+
+# MRD-EDGE Detection Agent
+
+The **MRD-EDGE Detection Agent** implements the MRD-EDGE (Enhanced Detection of ctDNA through Genomic Error suppression) deep learning algorithm for ultra-sensitive molecular residual disease detection. This AI-powered approach achieves unprecedented sensitivity in predicting cancer recurrence, detecting ctDNA at levels below 0.001% VAF with zero false negatives in validation studies.
+
+## When to Use This Skill
+
+* When standard ctDNA assays show negative but MRD is suspected.
+* For ultra-sensitive post-surgical MRD monitoring.
+* To detect relapse at the earliest possible timepoint.
+* When monitoring therapy response in minimal disease settings.
+* For research studies requiring highest sensitivity MRD detection.
+
+## Core Capabilities
+
+1. **Ultra-Sensitive Detection**: Detect ctDNA at 0.0001-0.001% VAF levels.
+
+2. **Deep Learning Error Suppression**: AI-powered sequencing error filtering.
+
+3. **Integrated Noise Modeling**: Patient-specific background noise estimation.
+
+4. **Multi-Feature Integration**: Combine mutations, fragmentation, methylation.
+
+5. **Zero False Negative Design**: Optimized for sensitivity while controlling specificity.
+
+6. **Longitudinal Tracking**: Monitor MRD over time with confidence intervals.
+
+## MRD-EDGE Algorithm Components
+
+| Component | Function | Improvement |
+|-----------|----------|-------------|
+| Error Suppression Network | Deep learning noise filter | 10x sensitivity |
+| Duplex Consensus | UMI-based error correction | 100x error reduction |
+| Fragment Analysis | Tumor fragment enrichment | 2-3x signal boost |
+| Integration Model | Multi-feature Bayesian fusion | Improved accuracy |
+
+## Sensitivity Comparison
+
+| Method | LOD (VAF) | False Negative Rate |
+|--------|-----------|---------------------|
+| Standard NGS | 1% | High |
+| UMI-corrected | 0.1% | Moderate |
+| Tumor-informed panels | 0.01% | Low |
+| MRD-EDGE | 0.001% | Near-zero |
+
+## Workflow
+
+1. **Input**: Deep sequenced cfDNA (>30,000x), tumor WES, matched normal.
+
+2. **Preprocessing**: UMI deduplication, duplex consensus, quality filtering.
+
+3. **Noise Modeling**: Patient-specific error profile estimation.
+
+4. **Feature Extraction**: Mutations, fragments, methylation signals.
+
+5. **Deep Learning Inference**: MRD-EDGE neural network prediction.
+
+6. **Bayesian Integration**: Combine features with uncertainty.
+
+7. **Output**: MRD probability, detected variants, confidence intervals.
+
+## Example Usage
+
+**User**: "Run MRD-EDGE analysis on this post-surgical colorectal cancer patient's plasma sample."
+
+**Agent Action**:
+```bash
+python3 Skills/Oncology/MRD_EDGE_Detection_Agent/mrd_edge_detect.py \
+    --cfdna_bam plasma_cfDNA.bam \
+    --tumor_vcf primary_tumor_mutations.vcf \
+    --normal_bam matched_normal.bam \
+    --coverage_depth 50000 \
+    --cancer_type colorectal \
+    --model_weights mrd_edge_v2.pt \
+    --output mrd_edge_results/
+```
+
+## Input Requirements
+
+| Input | Requirement | Purpose |
+|-------|-------------|---------|
+| cfDNA BAM | >30,000x depth, UMI-tagged | ctDNA detection |
+| Tumor VCF | WES/WGS mutations | Tumor-informed tracking |
+| Normal BAM | Matched germline | Background subtraction |
+| Coverage Depth | Minimum 30,000x | Sensitivity threshold |
+
+## Output Components
+
+| Output | Description | Format |
+|--------|-------------|--------|
+| MRD Probability | 0-1 probability of MRD | .json |
+| MRD Call | Positive/Negative with CI | .json |
+| Detected Variants | Variants contributing to call | .vcf |
+| Feature Scores | Per-feature contributions | .csv |
+| Noise Profile | Patient error model | .json |
+| Visualization | MRD landscape plot | .png |
+
+## Deep Learning Architecture
+
+| Layer | Function | Parameters |
+|-------|----------|------------|
+| Variant Encoder | Per-variant feature extraction | 2M |
+| Attention Layer | Cross-variant relationships | 1M |
+| Noise Classifier | Error vs true mutation | 5M |
+| Integration Head | Multi-feature fusion | 2M |
+| Output Layer | MRD probability | 100K |
+
+## Feature Categories
+
+| Category | Features | Weight |
+|----------|----------|--------|
+| Mutation Signal | VAF, read count, strand bias | Primary |
+| Fragment Features | Size, end motifs, coverage | Secondary |
+| Sequence Context | Trinucleotide, mappability | Noise correction |
+| Patient Background | Germline, CHIP, noise | Specificity |
+
+## Clinical Validation
+
+| Study | Cancer Type | Sensitivity | Specificity | Lead Time |
+|-------|-------------|-------------|-------------|-----------|
+| CRC Validation | Colorectal | 100% (5/5) | 95% | 10 months |
+| Lung Validation | NSCLC | 95% | 92% | 6 months |
+| Breast Validation | Breast | 93% | 94% | 12 months |
+
+## AI/ML Components
+
+**Error Suppression Network**:
+- Convolutional layers for sequence context
+- Recurrent layers for read-level features
+- Attention for cross-read patterns
+
+**Bayesian Integration**:
+- Prior from tumor mutational burden
+- Likelihood from detected signals
+- Posterior probability of MRD
+
+**Training Strategy**:
+- Semi-supervised with spike-in controls
+- Hard negative mining from CHIP
+- Transfer learning across cancer types
+
+## Prerequisites
+
+* Python 3.10+
+* PyTorch 2.0+
+* UMI-tools, fgbio for UMI processing
+* bcftools, samtools
+* MRD-EDGE model weights
+* High-memory compute (>64GB RAM)
+* GPU recommended
+
+## Related Skills
+
+* ctDNA_Dynamics_MRD_Agent - Longitudinal MRD tracking
+* Liquid_Biopsy_Analytics_Agent - Comprehensive liquid biopsy
+* CHIP_Clonal_Hematopoiesis_Agent - CHIP filtering
+* Tumor_Heterogeneity_Agent - Clonal tracking
+
+## Quality Control Metrics
+
+| Metric | Threshold | Interpretation |
+|--------|-----------|----------------|
+| Mean Coverage | >30,000x | Sensitivity adequate |
+| Duplex Rate | >20% | Error suppression possible |
+| cfDNA Input | >30ng | Sufficient material |
+| Tumor Mutations Tracked | >10 | Robust detection |
+| Background Noise | <0.001% | Specificity maintained |
+
+## Special Considerations
+
+1. **Sample Quality**: Requires high-quality cfDNA extraction
+2. **Sequencing Depth**: Deep sequencing essential for sensitivity
+3. **CHIP Exclusion**: Must filter clonal hematopoiesis variants
+4. **Tumor Heterogeneity**: Track clonal and subclonal mutations
+5. **Timing**: Sample >2 weeks post-surgery for clearance
+
+## Clinical Decision Support
+
+| MRD-EDGE Result | Recommended Action |
+|-----------------|-------------------|
+| MRD+ (high confidence) | Consider adjuvant therapy |
+| MRD+ (low confidence) | Repeat testing in 4-6 weeks |
+| MRD- (high confidence) | Surveillance per guidelines |
+| MRD- (low confidence) | Consider repeat testing |
+
+## Author
+
+AI Group - Biomedical AI Platform
+
+
+<!-- AUTHOR_SIGNATURE: 9a7f3c2e-MD-BABU-MIA-2026-MSSM-SECURE -->

package/data/skills/multi-ancestry-prs-agent/SKILL.md

@@ -0,0 +1,224 @@
+<!--
+# 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: 'multi-ancestry-prs-agent'
+description: 'AI-powered multi-ancestry polygenic risk score calculation and optimization for equitable disease risk prediction across diverse global populations.'
+measurable_outcome: Execute skill workflow successfully with valid output within 15 minutes.
+allowed-tools:
+  - read_file
+  - run_shell_command
+---
+
+
+# Multi-Ancestry PRS Agent
+
+The **Multi-Ancestry PRS Agent** provides AI-optimized polygenic risk score calculation designed to work across diverse ancestral populations. It addresses the critical limitation of European-biased GWAS by integrating trans-ancestry methods, improving risk prediction for underrepresented populations and enabling equitable precision medicine.
+
+## When to Use This Skill
+
+* When calculating PRS for non-European ancestry individuals.
+* For developing trans-ancestry risk prediction models.
+* To reduce PRS bias across ancestral populations.
+* When integrating multi-ancestry GWAS summary statistics.
+* For research on PRS portability and equity.
+
+## Core Capabilities
+
+1. **Multi-Ancestry PRS**: Calculate ancestry-aware polygenic scores.
+
+2. **Trans-Ancestry Optimization**: Optimize weights across populations.
+
+3. **Local Ancestry Integration**: Account for admixed genomes.
+
+4. **Ensemble Methods**: Combine multiple PRS approaches.
+
+5. **Ancestry Calibration**: Population-specific score calibration.
+
+6. **Equity Assessment**: Evaluate prediction fairness across groups.
+
+## Supported Ancestries
+
+| Ancestry | Training Data Availability | PRS Performance |
+|----------|---------------------------|-----------------|
+| European (EUR) | Excellent | High |
+| East Asian (EAS) | Good | Good |
+| South Asian (SAS) | Moderate | Moderate |
+| African (AFR) | Limited | Lower |
+| Hispanic/Latino (AMR) | Limited | Variable |
+| Middle Eastern (MID) | Very Limited | Lower |
+
+## Multi-Ancestry Methods
+
+| Method | Approach | Best For |
+|--------|----------|----------|
+| PRS-CSx | Cross-population shrinkage | Multi-ancestry |
+| PRS-Multi | Multi-population training | Large cohorts |
+| Ensemble | Combined methods | Best overall |
+| Local Ancestry | Ancestry-specific weights | Admixed populations |
+| GPSMult | Integrated multi-trait | Correlated traits |
+
+## Workflow
+
+1. **Input**: Individual genotypes, target ancestry, disease/trait.
+
+2. **Ancestry Inference**: Determine genetic ancestry.
+
+3. **Method Selection**: Choose optimal PRS approach.
+
+4. **Score Calculation**: Compute ancestry-aware PRS.
+
+5. **Calibration**: Apply population-specific calibration.
+
+6. **Risk Stratification**: Categorize into risk groups.
+
+7. **Output**: PRS, percentile, clinical interpretation.
+
+## Example Usage
+
+**User**: "Calculate multi-ancestry coronary artery disease PRS for this admixed individual with African and European ancestry."
+
+**Agent Action**:
+```bash
+python3 Skills/Precision_Medicine/Multi_Ancestry_PRS_Agent/calc_prs.py \
+    --genotypes patient_genotypes.vcf.gz \
+    --ancestry admixed_AFR_EUR \
+    --local_ancestry lai_segments.bed \
+    --trait coronary_artery_disease \
+    --method prs_csx \
+    --gwas_summary_stats eur_gwas.txt,afr_gwas.txt \
+    --calibration_cohort 1kg_admixed \
+    --output prs_results/
+```
+
+## Input Requirements
+
+| Input | Format | Purpose |
+|-------|--------|---------|
+| Genotypes | VCF/PLINK | Individual variants |
+| Ancestry | Estimated or self-reported | Method selection |
+| GWAS Summary Stats | Multiple ancestries | Score weights |
+| Local Ancestry | LAI segments | Admixture handling |
+| Reference Panel | Multi-ancestry | LD calculation |
+
+## Output Components
+
+| Output | Description | Format |
+|--------|-------------|--------|
+| PRS Score | Raw polygenic score | .csv |
+| Percentile | Population-specific ranking | .csv |
+| Risk Category | High/Intermediate/Low | .csv |
+| Ancestry Breakdown | Component scores | .json |
+| Confidence Interval | Score uncertainty | .json |
+| Clinical Interpretation | Risk explanation | .md |
+
+## Disease-Specific Performance
+
+| Disease | Multi-Ancestry AUC | EUR Only AUC | Improvement |
+|---------|-------------------|--------------|-------------|
+| CAD | 0.75-0.80 | 0.70-0.85 | 5-10% in non-EUR |
+| Type 2 Diabetes | 0.70-0.75 | 0.65-0.72 | 8-12% in AFR |
+| Breast Cancer | 0.65-0.72 | 0.60-0.70 | 5-8% globally |
+| Alzheimer's | 0.70-0.78 | 0.65-0.75 | 5-10% in diverse |
+
+## AI/ML Components
+
+**PRS Optimization**:
+- Bayesian shrinkage (PRS-CS)
+- Cross-population learning
+- Neural network weight optimization
+
+**Ancestry Inference**:
+- Supervised classification
+- Unsupervised clustering (PCA, ADMIXTURE)
+- Local ancestry inference (RFMix)
+
+**Ensemble Learning**:
+- Stacking multiple PRS methods
+- Ancestry-stratified weighting
+- Uncertainty quantification
+
+## Clinical Integration
+
+| Application | PRS Role | Clinical Action |
+|-------------|----------|-----------------|
+| Primary Prevention | Risk stratification | Screening intensity |
+| Risk Communication | Personalized risk | Lifestyle modification |
+| Treatment Selection | Predicted response | Drug choice |
+| Family Screening | Cascade testing | Genetic counseling |
+
+## Prerequisites
+
+* Python 3.10+
+* PLINK 2.0
+* PRSice-2, LDpred2, PRS-CSx
+* Multi-ancestry reference panels
+* GWAS summary statistics
+
+## Related Skills
+
+* PRS_Net_Deep_Learning_Agent - Deep learning PRS
+* Pharmacogenomics_Agent - Drug-gene interactions
+* PopEVE_Variant_Predictor_Agent - Variant interpretation
+* DiagAI_Agent - Clinical integration
+
+## Bias and Fairness
+
+| Bias Type | Cause | Mitigation |
+|-----------|-------|------------|
+| Discovery Bias | EUR-dominated GWAS | Multi-ancestry GWAS |
+| LD Variation | Population-specific LD | Local ancestry adjustment |
+| Allele Frequency | Differing frequencies | Population-specific weights |
+| Effect Size | Heterogeneous effects | Trans-ancestry meta-analysis |
+
+## Large-Scale Initiatives
+
+| Initiative | Focus | Contribution |
+|------------|-------|--------------|
+| All of Us | US diversity | 1M diverse participants |
+| PAGE | Multi-ethnic GWAS | Discovery in diverse |
+| H3Africa | African genomics | Continental diversity |
+| Mexican Biobank | Latin American | Admixed populations |
+| GBMI | Global Biobank | Multi-ancestry meta-analysis |
+
+## Special Considerations
+
+1. **Self-Reported Ancestry**: May not match genetic ancestry
+2. **Admixture**: Require local ancestry methods
+3. **Population Stratification**: Careful covariate adjustment
+4. **Clinical Validity**: Validate in target population
+5. **Health Equity**: Consider access disparities
+
+## ESC Guidelines Integration (2025)
+
+| Recommendation | PRS Role | Evidence Level |
+|----------------|----------|----------------|
+| CV Risk Assessment | Risk modifier | IIa, B |
+| Statin Decisions | Borderline risk reclassification | IIa, B |
+| Family History Enhancement | Quantify genetic burden | IIa, C |
+
+## Limitations
+
+| Limitation | Impact | Research Needed |
+|------------|--------|-----------------|
+| AFR Performance | Lower accuracy | More GWAS |
+| Rare Variants | Not captured | WGS integration |
+| Gene-Environment | Not modeled | Interaction studies |
+| Clinical Utility | Limited evidence | Randomized trials |
+
+## Author
+
+AI Group - Biomedical AI Platform
+
+
+<!-- AUTHOR_SIGNATURE: 9a7f3c2e-MD-BABU-MIA-2026-MSSM-SECURE -->