sevenTM empowers pharmaceutical and biotech companies to design and validate more candidates, in less time, with a focus on minimizing toxicity and side effects.
Our AI-powered tools, based on first principles, revolutionize the current drug discovery process for R&D teams with minimal disruption, while producing measurable improvements. Real-world validation through in-vitro feedback ensures the reliability and effectiveness of our models.
Our pipeline is particularly well suited to design therapeutics for low population diseases, especially those involving protein targets that are considered untargetable. We actively seek collaborations with private and academic partners for these projects.
Low population disease therapeutic design remains difficult, and without technological advance, these patients will remain underserved.
We have developed stand alone tools that augment and improve the early drug discovery process. This includes toxicity prediction, protein structure modeling, candidate docking and manufacturability prediction.
Our models gain real-world validation via in-silico/in-vitro feedback loops. We do this by partnering with R&D teams of Pharmaceutical and top researchers to design new candidates to move to clinical trials leveraging our AI-powered drug discovery pipeline.
Three orders of magnitude improvement over current high-throughput design and screening of small molecule drug candidates, accomplished through a cloud-native, end-to-end pipeline running on our cluster.
The top new candidates are tested in a highly automated fashion in our partners' labs.
LLM-assisted target and ligand identification
800-1200 pages analyzed in 5 minutes
Input required: user question + 1 click
Target modeling
Up to millisecond-scale Molecular Dynamics simulations
Input required: FASTA or PDB files + 1 click
Optional inputs: positive control(s), simulation parameters (force field, solvent model, membrane environment)
1 billion compounds screened per hour
1,000-10,000 compounds designed and ranked per hour
Input required: SMILES string(s) + 1 click
Optional input: library of compounds of interest. Default: ZINC20, comprising 800 million small molecules encoded by us in a format optimized for parallel computing
1,000-5,000 candidates analyzed per hour
Input required: tolerance threshold for toxicity + 1 click
Optional input: set or library of compounds of interest
1,000-5,000 compounds analyzed per hour
Input required: desired range from our scalability-stability tradeoff map + 1 click
Optional input: set or library of compounds of interest
Under construction
Virtually identified hits and designed lead compounds are assessed through a combination of assays that characterize their biophysics (target affinity) and cellular effects
This process is highly automated, with assays conducted in 96 and 384 plates
Quantum algorithms for protein molecular dynamics and docking
Personalized drug development and drug optimization based on genetic profiles
Automated drug discovery pipeline tailored for rare diseases