Molnovi
From protein target to ranked drug candidates
at computational scale and speed.
An open, modular pipeline that automates pocket detection, molecule generation, safety profiling, and binding estimation. You focus on the science.
The Problem
The Friction in Computational Chemistry
The computational workflow to find active molecules traditionally involves chaining together 5-6 separate open-source tools, each with different file formats, dependencies, and no shared infrastructure. Academic labs and biotechs do this manually with custom scripts. Things go wrong consistently.
- Wasted Cycles
-
Teams repeat experiments that have already been tried because there is no structured record of what was run, what failed, and why. There is no memory across campaigns.
- Silent Attrition
-
Molecules fail months later in synthesis or assay for reasons that were computationally detectable upfront. Toxicity flags, metabolic liabilities, things nobody checked because the pipeline didn't enforce it.
- Expert Bottleneck
-
Computational chemists spend 70%+ of their time managing file formats, job queues, and tool incompatibilities rather than analyzing molecules.
What We're Building
The Discovery Workflow
Input: a protein structure file (PDB). Output: a ranked shortlist of candidate drug molecules with predicted safety profiles, ready for a medicinal chemist to review.
Pocket Detection
ML-powered identification of druggable binding sites on the protein surface, validated against crystallographic ground truth.
Molecule Generation
Generates structurally diverse drug-like candidates, sized and shaped to fit the detected binding pocket.
Screening & Safety
Multi-layer drug-likeness filters followed by a 104-property toxicity, absorption, and metabolism profile for every candidate.
Docking
Physics-based binding affinity estimation, scoring how tightly each molecule fits the target pocket.
Reporting
Ranked candidates with per-molecule scorecards, safety profiles, explicit limitations, and full provenance tracking.
Validated Results
Pharmacological Validation
The pipeline independently finds molecules scoring in the same range as approved drugs, without any prior knowledge of those drugs.
EGFR (1M17)
Lung CancerKnown Drug: Erlotinib
Pocket Accuracy
2.7 Å from drug site
82% residue overlap
Best Docking Score
-9.32 kcal/mol
BCR-ABL (2HYY)
LeukemiaKnown Drug: Imatinib
Pocket Accuracy
2.7 Å from drug site
92% residue overlap
Best Docking Score
-12.56 kcal/mol
BRAF V600E (6P3D)
MelanomaKnown Drug: Ponatinib
Pocket Accuracy
3.1 Å from drug site
89% residue overlap
Best Docking Score
-10.40 kcal/mol
The pipeline independently finds molecules scoring in the same range as approved drugs, without any prior knowledge of those drugs. Every candidate is computationally profiled for drug-likeness, synthetic feasibility, and predicted safety before ranking.
What Molnovi is NOT
It is not a drug. It produces ranked hypotheses for experimental validation. It accelerates the work of computational chemists, but it does not replace their judgment.
We are honest about limitations: docking scores correlate with real binding at only r ≈ 0.4-0.6, ADMET predictions degrade on novel scaffolds, and every result must be validated experimentally.
Why It Matters
Our Approach
We are not competing with Schrödinger or Recursion. We are building accessible infrastructure for an underserved segment of a $2.5-5B market growing at 20-30% CAGR.
- Removing the $50K+ licensing barrier so any lab can run enterprise-grade pipelines without prohibitive software costs.
- Full Integration
- No more fragile pipelines. Stop chaining together six different tools with custom Python scripts that break on edge cases.
- Comprehensive Safety Profiling
- Every generated candidate immediately receives a 104-property toxicological and pharmacokinetic profile covering absorption, metabolism, toxicity, and more.
- Honest Limitations
- Every result includes confidence context and caveats, not just a score. We tell you when the data is weak or uncertain.
- Reproducible and auditable
- Every decision logged with full provenance tracking. No more lost parameters or untraceable runs.