molnovi.com is building an open, modular computational pipeline that automates the drug discovery workflow from protein structure to ranked candidate molecules. The platform integrates pocket detection, molecule generation, safety profiling, and binding estimation into a single system, eliminating the need for manual chaining of separate open-source tools. It is currently accepting pilot partners for early access.
Funding
Funding not disclosed
Founders
Product
Problem
Computational drug discovery traditionally requires chaining together 5-6 separate open-source tools, each with different file formats, dependencies, and no shared infrastructure, forcing academic labs and biotechs to manage workflows manually with custom scripts. This fragmented approach wastes cycles through repeated experiments with no structured record of what was run or failed, and molecules often fail months later in synthesis or assay due to toxicity flags and metabolic liabilities that were computationally detectable upfront but never checked. Computational chemists spend over 70% of their time managing file formats, job queues, and tool incompatibilities rather than analyzing molecules.
Solution
molnovi.com provides an open, modular pipeline that automates the entire discovery workflow from a protein structure file (PDB) input to a ranked shortlist of candidate drug molecules with predicted safety profiles. The platform integrates ML-powered pocket detection for identifying druggable binding sites, molecule generation, safety profiling, and binding estimation into a single cohesive system with shared infrastructure. By eliminating the need to manually chain separate tools, the pipeline enforces consistent checks for toxicity and metabolic liabilities upfront, reducing silent attrition. The system maintains a structured record of all experiments, creating institutional memory across campaigns so teams never repeat failed approaches. Computational chemists can focus on analyzing molecules rather than managing file formats and job queues, dramatically increasing productivity.
Target Audience
Primary users are computational chemists, medicinal chemists, and research teams in academic labs and biotech companies who need to streamline their drug discovery workflows and reduce time spent on tool management.
Features
- ML-powered pocket detection validated against crystallographic ground truth for identifying druggable binding sites
- Automated molecule generation with ranked output of candidate drug molecules
- Integrated safety profiling that flags toxicity and metabolic liabilities before synthesis
- Binding estimation capabilities built into the pipeline
- Structured experiment logging that creates searchable memory across campaigns
- Modular architecture allowing teams to use individual components or the full pipeline