C-MNY7physics-based AI, molecular modeling, drug discovery
0-100 chain-logic scale · 15 dimensions · scored on public evidence
Support predictive preclinical drug discovery using AI-accelerated, physics-based modeling.
Software platform for molecular modeling and drug discovery workflows.
Listed among Deep Origin's tools/platform offerings.
Develop or out-license a validated drug discovery asset targeting CD73.
AI-accelerated, physics-based drug discovery and validated-asset development.
Listed by Deep Origin as a validated asset/pipeline program.
Identify and prioritize candidate molecules for drug discovery.
Computational docking and virtual screening workflows within Deep Origin's AI/molecular-physics platform.
Listed as part of Deep Origin's platform capabilities.
Support patent-related analysis for drug discovery programs.
Software tool within Deep Origin's drug discovery platform suite.
Listed among Deep Origin's tools/platform offerings.
Provide a drug discovery software workspace for physics-based and AI-enabled modeling workflows.
SaaS platform access for computational drug discovery workflows.
Listed among Deep Origin's tools/platform offerings.
Estimate compound binding affinity changes to support drug discovery optimization.
Physics-based free-energy perturbation and relative binding free energy calculations.
A 2026 publication reported large-scale industrial assessment of OpenFE relative binding free energy calculations across more than 1,700 ligands.
The OpenFE protocol showed robust, reproducible performance with public-set weighted RMSE of about 1.73 kcal/mol and private-set weighted RMSE of about 2.44 kcal/mol, with some systems reaching sub-kcal/mol accuracy.
Develop or out-license a validated drug discovery asset targeting GPR75.
AI-accelerated, physics-based drug discovery and validated-asset development.
Listed by Deep Origin as a validated asset/pipeline program.
Replace animal testing with in-silico models for predictive preclinical development.
AI-accelerated, physics-based in-silico modeling under an ARPA-H-funded program.
Deep Origin received a $31.7M ARPA-H contract in 2026 for this program.
Create large molecular conformation and property datasets to accelerate drug discovery and improve molecular energy models.
Volunteer distributed computing for quantum chemistry DFT calculations combined with active learning, molecular dynamics, and machine learning model ensembles.
Published in Scientific Reports in 2025 after a 2024 bioRxiv preprint.
Generated a public dataset of diverse ENAMINE molecules with calculated energies and benchmarked ensemble ML models for molecular energy prediction.