Products
Find the right model or tool for your drug discovery work.
AI/ML models and tools available now on TuneLab
AbLab v1, v2
Predicts antibody developability across 22 endpoints, including stability, aggregation, and solubility, from sequence alone.
Antibody-Antigen Binding Affinity (ΔΔG)
TuneLab’s Antibody–Antigen Binding Affinity model predicts how a specific mutation changes binding affinity between an antibody and its antigen.
ChemLab v1, v2 Federated
Predicts ADMET and physicochemical properties to help prioritize compounds for synthesis.
Immunogenicity Model (MHC-II Peptide Presentation)
Predicts MHC class II presentation likelihood and 9-mer binding cores for antibody peptides across 12 HLA-DRB1 alleles.
in vivo PK (developed by insitro)
Predicts preclinical in vivo PK, including clearance and Vss, across multi species, developed by insitro for Lilly.
NanoLab (VHH Developability)
Predicts five biophysical developability endpoints for VHH nanobody sequences to rank leads before wet lab testing.
Small Molecule Computational Workflows
Configurable workflows that generate, filter, score, predict properties and rank small molecules to automate drug discovery campaigns.
TuneLab’s small molecule computational workflows are configurable, automated pipelines that connect multiple steps in a drug discovery campaign. A workflow can generate or retrieve molecules, filter unwanted structures, predict properties, and rank candidates, eliminating the need to run each tool separately and combine results manually.
There are currently two pre-configured workflows:
- Hit Expansion: Starting with one or more confirmed active compounds, this workflow searches vendor screening libraries for similar molecules and selects new candidates based on similarity threshold. Multiple searches can run in parallel, after which TuneLab automatically combines and deduplicates the results. The resulting molecules can then be filtered, scored using predicted properties, and ranked according to project-specific criteria.
- Generative Molecular Design: Starting with known fragments or scaffolds, this workflow uses NVIDIA BioNeMo’s GenMol model to generate new molecular structures. The generated molecules are evaluated using structural filters, enabling users to filter out the proposed molecules where the structure either has a known common liability or is not represented in ChEMBL. Molecules that pass the selected filters can then be ranked based on predicted properties.
Tools
- 2D Similarity Search: Find close analogs of known compounds in the vendor screening libraries.
- Lilly Search: Lilly’s proprietary molecular search engine and fingerprinting technology.
- Arthor by NextMove Software: A high-performance similarity-search tool.
- NVIDIA BioNeMo nvMolKit¹: A BioNeMo Python library that accelerates common molecular-search operations.
- Generative Chemistry Models: Propose new molecular structures using NVIDIA BioNeMo GenMol.²
- Structural Filters: Remove or flag potentially undesirable structures.
- Substructure Filter: Built on top of RDKit that applies structural alerts and property filters to chemical structures.³
- Ring Precedence Filter: A Lilly-developed filter that flags molecules containing ring systems not found in ChEMBL.
- Property Prediction: Use ChemLab models to predict commonly measured in vitro ADME properties.
- 2D Similarity Search: Find close analogs of known compounds in the vendor screening libraries.
Features
- Flexible filtering: Apply thresholds to any property or score calculated earlier in the workflow.
- Live progress tracking: See which step is currently running and how long each completed step took.
- Reusable templates: Save a trusted workflow configuration and reuse it as the starting point for future campaigns.
Modality
The current workflows are designed for small-molecule drug discovery. Antibody discovery workflows are not supported in this release.
- ¹https://research.nvidia.com/labs/dbr/blog/nvMolKit/
- ²https://build.nvidia.com/nvidia/genmol-generate
- ³https://github.com/PatWalters/rd_filters