@bgicli/bgicli 2.1.1 → 2.2.1

package/data/skills/cryoem-ai-drug-design-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: 'cryoem-ai-drug-design-agent'
+description: 'AI-powered integration of cryo-EM structural data with generative AI and molecular dynamics for structure-based drug design targeting flexible proteins and membrane complexes.'
+measurable_outcome: Execute skill workflow successfully with valid output within 15 minutes.
+allowed-tools:
+  - read_file
+  - run_shell_command
+---
+
+
+# Cryo-EM AI Drug Design Agent
+
+The **Cryo-EM AI Drug Design Agent** integrates cryo-electron microscopy structural data with AlphaFold3, generative AI, and molecular dynamics for structure-based drug design. It enables targeting of previously "undruggable" proteins including flexible, membrane-bound, and large macromolecular complexes through high-resolution structure-guided optimization.
+
+## When to Use This Skill
+
+* When designing drugs against cryo-EM-solved targets.
+* For fragment-based drug discovery with EM structures.
+* To model ligand binding in flexible protein regions.
+* When targeting membrane proteins and large complexes.
+* For integrating AlphaFold predictions with experimental EM density.
+
+## Core Capabilities
+
+1. **Density-Guided Design**: Fit ligands into cryo-EM density maps.
+
+2. **AlphaFold Integration**: Combine AF3 predictions with EM data.
+
+3. **Flexible Docking**: Account for protein dynamics in binding.
+
+4. **Fragment Screening**: Virtual fragment screening with EM structures.
+
+5. **Complex Targeting**: Design for multi-protein assemblies.
+
+6. **Dynamics-Based Design**: Incorporate conformational flexibility.
+
+## Cryo-EM for Drug Discovery
+
+| Target Class | Cryo-EM Advantage | Drug Discovery Application |
+|--------------|-------------------|---------------------------|
+| GPCRs | Native lipid environment | Allosteric sites |
+| Ion Channels | Multiple conformations | State-specific design |
+| Transporters | Conformational states | Mechanism-based |
+| Ribosomes | Antibiotic binding | New antibiotics |
+| Viral Proteins | Large assemblies | Vaccines, antivirals |
+| Intrinsically Disordered | Flexible regions | Challenging targets |
+
+## Workflow
+
+1. **Input**: Cryo-EM density map, protein sequence, ligand/fragment.
+
+2. **Structure Refinement**: AlphaFold + density-guided refinement.
+
+3. **Binding Site Identification**: Detect pockets in EM structure.
+
+4. **Ligand Placement**: Density-guided ligand fitting.
+
+5. **MD Simulation**: Flexible binding simulation.
+
+6. **Optimization**: Generative design around hits.
+
+7. **Output**: Optimized ligands, binding models, design recommendations.
+
+## Example Usage
+
+**User**: "Design ligands for this GPCR cryo-EM structure, accounting for receptor flexibility in the binding pocket."
+
+**Agent Action**:
+```bash
+python3 Skills/Structural_Biology/CryoEM_AI_Drug_Design_Agent/design_from_cryoem.py \
+    --density_map gpcr_3.2A.mrc \
+    --protein_sequence gpcr.fasta \
+    --alphafold_model gpcr_af2.pdb \
+    --resolution 3.2 \
+    --ligand_screening fragment_library.sdf \
+    --binding_site_residues "3.32,5.46,6.48,7.39" \
+    --md_refinement true \
+    --generative_optimization true \
+    --output gpcr_drug_design/
+```
+
+## Input Requirements
+
+| Input | Format | Purpose |
+|-------|--------|---------|
+| Density Map | MRC/MAP | EM density |
+| Protein Sequence | FASTA | AlphaFold input |
+| Resolution | Float (Å) | Quality metric |
+| Ligand Library | SDF | Virtual screening |
+| Known Ligand | Optional SDF | Starting point |
+
+## Output Components
+
+| Output | Description | Format |
+|--------|-------------|--------|
+| Refined Structure | EM + AF combined | .pdb |
+| Ligand Poses | Density-fitted poses | .sdf |
+| Binding Scores | Affinity predictions | .csv |
+| Optimized Compounds | Generative designs | .sdf |
+| MD Trajectory | Flexibility analysis | .xtc |
+| Design Report | Recommendations | .pdf |
+
+## AI/ML Components
+
+**Structure Prediction**:
+- AlphaFold3 for initial model
+- Density-guided refinement
+- Confidence scoring (pLDDT, local resolution)
+
+**Ligand Design**:
+- Generative AI (diffusion, VAE)
+- Reinforcement learning optimization
+- Multi-objective scoring
+
+**Dynamics Integration**:
+- Molecular dynamics simulation
+- Ensemble docking
+- Flexibility-aware scoring
+
+## Resolution Considerations
+
+| Resolution | Applications | Limitations |
+|------------|--------------|-------------|
+| <3.0 Å | Fragment screening, detailed design | Rare |
+| 3.0-4.0 Å | Drug optimization, binding mode | Most targets |
+| 4.0-5.0 Å | Pocket identification, scaffold | Less detail |
+| >5.0 Å | Architecture, general binding | Low for SBDD |
+
+## AlphaFold3 + Cryo-EM Integration
+
+| Scenario | Approach | Benefit |
+|----------|----------|---------|
+| Missing Loops | AF3 prediction | Complete structure |
+| Flexible Regions | Ensemble models | Multiple conformations |
+| Low Resolution | AF3 template | Higher confidence |
+| Ligand Binding | AF3 complex prediction | Binding mode |
+
+## Prerequisites
+
+* Python 3.10+
+* AlphaFold3, ChimeraX
+* GROMACS/OpenMM for MD
+* RDKit, AutoDock Vina
+* GPU with 16GB+ VRAM
+
+## Related Skills
+
+* Time_Resolved_CryoEM_Agent - Dynamics from EM
+* PROTAC_Design_Agent - Degrader design
+* Molecular_Glue_Discovery_Agent - Glue design
+* AlphaFold3_Agent - Structure prediction
+
+## Fragment-Based Discovery with Cryo-EM
+
+| Step | Method | Cryo-EM Role |
+|------|--------|--------------|
+| Fragment Screening | Virtual dock to EM | Density-guided |
+| Hit Identification | Cryo-EM soaking | Experimental validation |
+| Fragment Growing | EM + modeling | Structure guidance |
+| Lead Optimization | Iterative EM | Binding mode confirmation |
+
+## Membrane Protein Targets
+
+| Target Type | Cryo-EM Advantage | Examples |
+|-------------|-------------------|----------|
+| GPCRs | Native membrane | Numerous drugs |
+| Ion Channels | State-dependent | Painkillers, antiepileptics |
+| Transporters | Mechanism insight | Cancer, infection |
+| Receptors | Complex structures | Immunotherapy |
+
+## Special Considerations
+
+1. **Resolution Limits**: Design confidence depends on resolution
+2. **Map Quality**: Local resolution varies across structure
+3. **Conformational States**: Multiple states may be captured
+4. **Ligand Density**: May be weak at lower resolution
+5. **Validation**: Experimental validation essential
+
+## Quality Metrics
+
+| Metric | Purpose | Threshold |
+|--------|---------|-----------|
+| Global Resolution | Overall quality | <4.0 Å for SBDD |
+| Local Resolution | Binding site quality | <3.5 Å preferred |
+| Map Correlation | Model-to-map fit | >0.8 |
+| Real-Space R | Atomic fit | <0.3 |
+| Ligand CCC | Ligand fit | >0.6 |
+
+## Drug Discovery Success Stories
+
+| Drug | Target | Cryo-EM Role |
+|------|--------|--------------|
+| Numerous | GPCRs | Structure determination |
+| Antibiotics | Ribosome | Binding mode |
+| Antivirals | Spike protein | Epitope mapping |
+| Various | Ion channels | State-specific design |
+
+## Author
+
+AI Group - Biomedical AI Platform
+
+
+<!-- AUTHOR_SIGNATURE: 9a7f3c2e-MD-BABU-MIA-2026-MSSM-SECURE -->

package/data/skills/ctdna-dynamics-mrd-agent/SKILL.md

@@ -0,0 +1,206 @@
+<!--
+# 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: 'ctdna-dynamics-mrd-agent'
+description: 'AI-powered circulating tumor DNA dynamics analysis for molecular residual disease detection, treatment response monitoring, and early relapse prediction using liquid biopsy.'
+measurable_outcome: Execute skill workflow successfully with valid output within 15 minutes.
+allowed-tools:
+  - read_file
+  - run_shell_command
+---
+
+
+# ctDNA Dynamics MRD Agent
+
+The **ctDNA Dynamics MRD Agent** provides comprehensive analysis of circulating tumor DNA dynamics for molecular residual disease (MRD) detection, treatment response monitoring, and early relapse prediction. It integrates tumor-informed and tumor-naive approaches with temporal modeling for longitudinal ctDNA analysis.
+
+## When to Use This Skill
+
+* When monitoring minimal/molecular residual disease post-treatment.
+* For tracking treatment response through ctDNA kinetics.
+* To predict relapse before clinical/radiological detection.
+* When assessing tumor burden dynamics during therapy.
+* For early detection of acquired resistance mutations.
+
+## Core Capabilities
+
+1. **MRD Detection**: Ultra-sensitive detection of residual disease (LOD 0.001% VAF).
+
+2. **Kinetic Modeling**: Model ctDNA clearance and doubling time.
+
+3. **Response Prediction**: Predict treatment response from early ctDNA dynamics.
+
+4. **Relapse Prediction**: Identify molecular relapse months before imaging.
+
+5. **Resistance Monitoring**: Track emergence of resistance mutations.
+
+6. **Multi-Timepoint Integration**: Analyze longitudinal ctDNA trajectories.
+
+## Detection Approaches
+
+| Approach | Method | LOD | Best Use Case |
+|----------|--------|-----|---------------|
+| Tumor-Informed | Track known mutations | 0.001% | Post-surgical MRD |
+| Tumor-Naive | Panel-based detection | 0.1% | Screening, unknown primary |
+| WGS-Based | Fragmentomics + mutations | 0.01% | Comprehensive profiling |
+| Methylation | cfDNA methylation | 0.1% | Tissue of origin, early detection |
+
+## Kinetic Parameters
+
+| Parameter | Definition | Clinical Meaning |
+|-----------|------------|------------------|
+| ctDNA Half-Life | Time to 50% reduction | Treatment sensitivity |
+| Doubling Time | Time to 2x increase | Tumor growth rate |
+| Nadir | Lowest ctDNA level | Depth of response |
+| Time to Nadir | Days to reach nadir | Response kinetics |
+| Clearance Rate | Exponential decay constant | Treatment efficacy |
+| Lead Time | MRD+ to clinical relapse | Early detection window |
+
+## Workflow
+
+1. **Input**: Serial ctDNA measurements (VAF or copies/mL), timepoints, treatment dates.
+
+2. **QC**: Assess sequencing quality, coverage, tumor fraction.
+
+3. **Mutation Tracking**: Quantify tracked variants across timepoints.
+
+4. **Kinetic Modeling**: Fit exponential/sigmoidal models to dynamics.
+
+5. **MRD Calling**: Determine MRD status with confidence intervals.
+
+6. **Resistance Detection**: Identify emerging resistant clones.
+
+7. **Output**: MRD status, kinetic parameters, predictions, visualizations.
+
+## Example Usage
+
+**User**: "Analyze this patient's serial ctDNA data to assess MRD status and predict relapse risk."
+
+**Agent Action**:
+```bash
+python3 Skills/Oncology/ctDNA_Dynamics_MRD_Agent/ctdna_mrd_analysis.py \
+    --ctdna_data serial_ctdna.tsv \
+    --tracked_mutations tumor_mutations.vcf \
+    --sample_times 0,14,42,90,180 \
+    --treatment_start 0 \
+    --surgery_date 7 \
+    --cancer_type colorectal \
+    --output mrd_analysis/
+```
+
+## Input Data Format
+
+```tsv
+Sample_ID  Timepoint_Days  Mutation  VAF  Copies_per_mL  Coverage
+PT001_T0   0               TP53_R248Q  5.2  1500          15000
+PT001_T1   14              TP53_R248Q  2.1  620           18000
+PT001_T2   42              TP53_R248Q  0.05 15            20000
+PT001_T3   90              TP53_R248Q  0.002 0.6          22000
+```
+
+## Output Components
+
+| Output | Description | Format |
+|--------|-------------|--------|
+| MRD Status | Positive/Negative at each timepoint | .csv |
+| Kinetic Parameters | Half-life, doubling time, nadir | .json |
+| Response Classification | Major/Minor/No response | .csv |
+| Relapse Risk | Probability and predicted time | .json |
+| Dynamics Plot | ctDNA trajectory visualization | .png, .pdf |
+| Resistance Variants | Emerging mutations | .vcf |
+| Clonal Evolution | Clone frequency over time | .csv |
+
+## Response Definitions
+
+| Response Category | ctDNA Change | Clinical Correlation |
+|-------------------|--------------|---------------------|
+| Major Molecular Response | >2 log reduction | Excellent prognosis |
+| Molecular Response | 1-2 log reduction | Good prognosis |
+| Stable Molecular Disease | <1 log change | Intermediate |
+| Molecular Progression | >0.5 log increase | Poor prognosis |
+
+## Cancer-Specific Parameters
+
+| Cancer Type | Typical Half-Life | MRD Lead Time | ctDNA Shedding |
+|-------------|-------------------|---------------|----------------|
+| Colorectal | 1-2 days | 6-12 months | High |
+| Lung (NSCLC) | 1-3 days | 3-6 months | High |
+| Breast | 2-5 days | 6-18 months | Moderate |
+| Pancreatic | 1-2 days | 3-6 months | High |
+| Melanoma | 2-4 days | 3-9 months | Variable |
+
+## AI/ML Components
+
+**Kinetic Modeling**:
+- Non-linear mixed effects models
+- Bayesian hierarchical models
+- Gaussian process regression
+
+**MRD Detection**:
+- Error-suppressed variant calling
+- Machine learning noise filtering
+- Duplex UMI deduplication
+
+**Relapse Prediction**:
+- Time-series forecasting (LSTM, Transformers)
+- Survival analysis (Cox, Random Survival Forests)
+- Multi-mutation integration
+
+## Clinical Trial Support
+
+| Application | Endpoint | ctDNA Metric |
+|-------------|----------|--------------|
+| Neoadjuvant | pathCR surrogate | Pre-surgery clearance |
+| Adjuvant | DFS surrogate | Post-surgery MRD |
+| Metastatic | PFS/OS surrogate | ctDNA dynamics |
+| Maintenance | Duration decision | MRD negativity |
+
+## Prerequisites
+
+* Python 3.10+
+* Variant callers (Mutect2, Strelka)
+* UMI-aware pipelines
+* scipy, lifelines, survival analysis tools
+* PyTorch for deep learning models
+
+## Related Skills
+
+* MRD_EDGE_Detection_Agent - Ultra-sensitive MRD detection
+* Liquid_Biopsy_Analytics_Agent - Comprehensive liquid biopsy
+* Tumor_Heterogeneity_Agent - Clonal evolution tracking
+* HRD_Analysis_Agent - Genomic biomarkers
+
+## Special Considerations
+
+1. **Tumor Fraction**: Low tumor fraction limits sensitivity
+2. **Pre-Analytical**: Plasma processing affects cfDNA quality
+3. **Clonal Hematopoiesis**: CHIP variants can confound results
+4. **Panel Design**: Ensure sufficient mutation coverage
+5. **Timing**: Sample timing relative to treatment critical
+
+## FDA-Cleared ctDNA Tests
+
+| Test | Cancer Types | Application |
+|------|--------------|-------------|
+| Guardant360 CDx | Pan-cancer | Treatment selection |
+| FoundationOne Liquid CDx | Pan-cancer | Treatment selection |
+| Signatera | Solid tumors | MRD monitoring |
+| Guardant Reveal | CRC | MRD detection |
+
+## Author
+
+AI Group - Biomedical AI Platform
+
+
+<!-- AUTHOR_SIGNATURE: 9a7f3c2e-MD-BABU-MIA-2026-MSSM-SECURE -->

package/data/skills/cytokine-storm-analysis-agent/SKILL.md

@@ -0,0 +1,180 @@
+<!--
+# 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: 'cytokine-storm-analysis-agent'
+description: 'AI-powered cytokine release syndrome (CRS) and cytokine storm analysis for prediction, monitoring, and management in immunotherapy and infectious disease.'
+measurable_outcome: Execute skill workflow successfully with valid output within 15 minutes.
+allowed-tools:
+  - read_file
+  - run_shell_command
+---
+
+
+# Cytokine Storm Analysis Agent
+
+The **Cytokine Storm Analysis Agent** provides comprehensive AI-driven analysis of cytokine release syndrome (CRS) and hyperinflammatory states. It integrates cytokine profiling, clinical parameters, and immunological markers for early prediction, severity grading, and treatment guidance in CAR-T therapy, sepsis, and viral infections.
+
+## When to Use This Skill
+
+* When monitoring CAR-T patients for cytokine release syndrome risk.
+* To predict CRS severity and timing post-immunotherapy.
+* For analyzing cytokine panels in sepsis and viral infections (COVID-19).
+* When guiding tocilizumab/siltuximab anti-IL-6 therapy decisions.
+* To distinguish CRS from ICANS, HLH, and other inflammatory syndromes.
+
+## Core Capabilities
+
+1. **CRS Risk Prediction**: ML models predict CRS development and severity from baseline factors (tumor burden, disease type, CAR-T product).
+
+2. **Real-Time Monitoring**: Track cytokine dynamics (IL-6, IFN-γ, IL-10, ferritin) with early warning alerts.
+
+3. **Severity Grading**: Automated ASTCT CRS grading using clinical parameters and biomarkers.
+
+4. **Differential Diagnosis**: Distinguish CRS from HLH/MAS, ICANS, infection, and tumor lysis syndrome.
+
+5. **Treatment Guidance**: AI-driven recommendations for tocilizumab, corticosteroids, and supportive care.
+
+6. **Outcome Prediction**: Model response to anti-cytokine therapy and overall outcomes.
+
+## Cytokine Panel Analysis
+
+| Cytokine | Role in CRS | Kinetics | Therapeutic Target |
+|----------|-------------|----------|-------------------|
+| IL-6 | Central mediator | Early peak | Tocilizumab, Siltuximab |
+| IFN-γ | T-cell activation | Early | Emapalumab |
+| IL-1β | Inflammasome | Early | Anakinra |
+| IL-10 | Regulatory | Variable | - |
+| TNF-α | Pro-inflammatory | Early | Infliximab (caution) |
+| IL-2 | T-cell expansion | Early | - |
+| GM-CSF | Myeloid activation | Sustained | Lenzilumab |
+
+## ASTCT CRS Grading (Automated)
+
+| Grade | Fever | Hypotension | Hypoxia |
+|-------|-------|-------------|---------|
+| 1 | ≥38°C | None | None |
+| 2 | ≥38°C | Responsive to fluids | Low-flow O2 |
+| 3 | ≥38°C | One vasopressor | High-flow O2 |
+| 4 | ≥38°C | Multiple vasopressors | Ventilation |
+
+## Workflow
+
+1. **Input**: Cytokine levels, vital signs, laboratory values, treatment history.
+
+2. **Risk Assessment**: Baseline CRS risk stratification pre-therapy.
+
+3. **Monitoring**: Real-time cytokine tracking with trend analysis.
+
+4. **Grading**: Automated CRS grade assignment per ASTCT criteria.
+
+5. **Differential**: Rule out mimics (infection, HLH, ICANS).
+
+6. **Treatment**: Generate management recommendations.
+
+7. **Output**: CRS risk score, grade, differential diagnosis, treatment plan.
+
+## Example Usage
+
+**User**: "Monitor this CAR-T patient's cytokine levels and predict CRS severity."
+
+**Agent Action**:
+```bash
+python3 Skills/Immunology_Vaccines/Cytokine_Storm_Analysis_Agent/crs_analyzer.py \
+    --patient_data demographics.json \
+    --cytokines cytokine_panel.csv \
+    --vitals vital_signs.csv \
+    --labs laboratory_values.csv \
+    --cart_product tisagenlecleucel \
+    --day_post_infusion 5 \
+    --model crs_predictor_v3 \
+    --output crs_report.json
+```
+
+## AI/ML Models
+
+**CRS Risk Prediction**:
+- Features: tumor burden (LDH), lymphodepletion intensity, CAR-T dose, disease type
+- Model: Gradient boosting with SHAP interpretability
+- Performance: AUC 0.82-0.88 for severe CRS
+
+**Severity Trajectory**:
+- Time-series modeling of cytokine dynamics
+- LSTM networks for temporal patterns
+- Early warning 24-48 hours before clinical deterioration
+
+**Treatment Response**:
+- Tocilizumab response prediction
+- Corticosteroid escalation timing
+- ICU admission risk
+
+## Differential Diagnosis Decision Tree
+
+```
+Fever + Elevated Cytokines
+          |
+    CAR-T context?
+    /           \
+  Yes            No
+   |              |
+Hypotension?   Infection workup
+   |              |
+  CRS          Sepsis vs viral
+   |
+Neuro symptoms?
+   |
+  ICANS vs CRS
+   |
+Ferritin >10,000?
+   |
+  HLH/MAS evaluation
+```
+
+## Clinical Decision Support
+
+**Tocilizumab Indication**:
+- Grade 2+ CRS
+- Rapidly rising cytokines
+- High-risk baseline features
+
+**Corticosteroid Indication**:
+- Tocilizumab-refractory CRS
+- ICANS any grade
+- Grade 3+ CRS
+
+## Prerequisites
+
+* Python 3.10+
+* scikit-learn, XGBoost for ML
+* Time-series analysis libraries
+* FHIR client for EHR integration
+
+## Related Skills
+
+* CART_Design_Optimizer_Agent - For CAR-T design
+* TCell_Exhaustion_Analysis_Agent - For T-cell function
+* Clinical_NLP - For extracting symptoms from notes
+
+## Special Populations
+
+1. **Pediatric**: Different baseline cytokine ranges
+2. **Post-COVID**: Altered inflammatory responses
+3. **Bridging Therapy**: Impact on CRS risk
+4. **Concurrent Infection**: Confounding cytokine elevation
+
+## Author
+
+AI Group - Biomedical AI Platform
+
+
+<!-- AUTHOR_SIGNATURE: 9a7f3c2e-MD-BABU-MIA-2026-MSSM-SECURE -->