The DNAI Pipeline Explorer
Interactive visualization of our complete multi-omics to evolutionary dynamics pipeline. Click any model for detailed architecture, metrics, and specifications.
Interactive Pipeline Diagram
Hover over models to see connections. Click to explore detailed specifications.
All Models
Deep dive into each component of the DNAI pipeline
Multi-modal VAE
Compresses a patient's full molecular profile into a biological fingerprint
Domain Separation Network
Translates mouse model data into human-relevant predictions
CausalDriver-GAT
Identifies which mutations are actually driving the cancer
TxResponse
Predicts how sensitive a tumor is to different drugs
Hypernetwork
Generates personalized tumor parameters for each patient
Neural ODE
Simulates how a tumor responds to treatment over time
EvoSim
ODE-coupled stochastic ensemble for clonal evolution under treatment
Treatment Optimizer
Ranks treatment regimens using counterfactual reasoning
Synthetic Lethality Engine
Identifies druggable vulnerabilities from mutation-pathway interactions
Immunogenic Variant Candidates
Prioritizes variants for immunotherapy based on tumor microenvironment
Methylation Decoder
Reconstructs epigenetic patterns and detects silenced tumor suppressors
Combination Discovery Engine
Predicts synergistic drug combinations from monotherapy data alone
End-to-End Data Flow
From raw multi-omics data to evolutionary predictions in one unified pipeline
Multi-Omics Input
- RNA-seq (2,579 genes)
- CNV (1,886 genes)
- Mutations (500 genes)
- Methylation (1,000 probes)
Foundation Model
VAE encodes to 328-dimensional latent space with biological structure and interpretability
Dual Pipeline
V1 static analysis + V2 dynamic simulation working in parallel for comprehensive insights
Predictions
- Driver mutations
- Drug sensitivity
- Clone trajectories
- Evolution risk
Ready to explore the details?
Click any model above to see full architecture specifications, mathematical formulations, and validation metrics.