Seeking Strategic Partnerships
Spektron Systems seeks a select few strategic partners to collaboratively deploy Q-MAP’s powerful, AI-based Drug Discovery capacity.
- Novel lead generation from data associating outcomes with structures
- Moiety and sub-moiety causal inference with efficacy, toxicity, and ADMET outcomes
- De-risking of leads through sales/marketing via multi-parameter optimization
- Predictive ADMET modeling and off-target identification
- ID of candidates for drug repurposing
Extending our collaborative work with the National Center for Toxicological Research, Spektron Systems has developed Q-MAP™ – a platform for AI-assisted, target-based drug design. This platform designs best-in-class novel molecules at 25% of the cost and time of conventional drug discovery.
AI for Target-based Drug Design
Fast, Iterative Drug Design
The Q-MAP™ platform takes a multi-model approach to reduce uncertainty in predictions for clinical results; modeling across multiple endpoints from proposed mechanism of action to measured clinical metric. With active moiety inference and predictive ADMET models, Q-MAP™ can generate novel molecules optimized for multiple parameters before synthesis. By prioritizing in silico leads and de-risking them before synthesis, we can reduce costs and shorten discovery timelines from biology to preclinical candidates (PCC) from 4.5+ years to approximately 1 year and to Investigational New Drug (IND) within 3.5 years.
Q-MAP™ models multiple endpoints, both on-target and off-target, to pre-optimize molecule design and identify related structural moieties.
Novel Compounds Generation
Our approach enables an informed building block method for generating a library of novel molecules enriched by modeling on relevant endpoints.
In a rapid and lean Design-Make-Test-Analyze approach, Q-MAP™ actively learns from results as molecules are designed, synthesized, and tested physically against a key profile.
The Q-MAP™ platform, not based on chemical structures, is able to create novel, potent drug candidates which are structurally unrelated to the existing drugs and, simultaneously have adequate or superior biological properties, including receptor affinities.
Tractable qHTS Data Analysis
Qualitative data meets quantum data
Q-MAP™ methodology is tractable without expensive computational costs. This allows for fast analysis of large bespoke data sets generated from qHTS assays. Our methods identify active moieties related to custom endpoints and help establish a vocabulary for building novel molecules.
Predictive Modeling and Virtual Screening
Machine learning applied to curated, small-molecule data
Data - Big and Small
The world of chemistry and pharmaceutical development generates many data sets. Some are Big Data and some are small data sets. Being able to work with small datasets where Big Data algorithms fail is crucial to modeling against endpoints where tests are invasive or costly.
Q-MAP’s predictive modeling is based on applied machine learning theory combined expertise in computational chemistry, pharmacology, and software development.
Q-MAP is built on tractable learning methods using best practices in applied machine learning, software engineering, and computational chemistry.
Dr. David Wolf will be speaking on a panel presentation about AI in Healthcare and Spektron’s unique approach to medicinal molecule design. This invite-only presentation will be attended by investors looking to learn more about AI in this industry. Hosted by IMPACTIVE BioConsult, this presentation appends the 2019 BIO CEO SUMMIT and is being held 7:00 pm at Morgan Stanley.
Dr. David Wolf, the President and Chief Medical Officer for Spektron Systems, sat on the panel with other distinguished speakers talking about Artificial Intelligence. In this backstage interview clip, he about the revolutionary impact that AI is having in many industries. He addresses one of the key components of any good machine learning implementation – “Curation! Curation! Curation!”
417 Main St, Suite 129
Little Rock, AR 72201
Visit us on LinkedIn or follow us on Twitter and Facebook!