News & Events

The promise of AI in drug discovery is not just about finding novel molecules; it’s about finding viable ones. The pharmaceutical industry continues to face high failure rates, often due to poor Absorption, Distribution, Metabolism, Excretion, and Toxicity (ADMET) properties identified late in the development cycle.

Today, we are thrilled to announce a strategic partnership with Expert Systems, Inc. to launch Q-Models, a powerful new offering within the Quantori Q-Suite platform. Q-Models is designed to mitigate these risks by bringing over 175 state-of-the-art predictive ML models directly to your fingertips, integrated seamlessly into your R&D workflows.

The Cost of Late-Stage Failure

ADMET testing is the crucial gatekeeper in drug development. A compound may show perfect efficacy in vitro, but if it fails to be absorbed in the gut (Absorption), is rapidly broken down by the liver (Metabolism), or proves toxic (Toxicity), it is destined to fail. Finding these flaws late is prohibitively expensive and wastes years of effort.

Q-Models tackles this challenge head-on by providing high-quality, pre-validated ADMET and physicochemical predictions early in the discovery phase.

Q-Models: Intelligence Built for Life Sciences

Q-Models is unique in that it is delivered not as a standalone tool, but as a core service integrated into the modular Q-Suite architecture (Q-Discover, Q-Scientist, and Q-Flow). This deep integration ensures predictions are contextually relevant and immediately actionable within your AI-driven molecular design and screening protocols.

Key Advantages of Q-Models:

1. Comprehensive, Curated Models
   The offering includes a vast library of over 175 models covering everything from Caco-2 permeability and Solubility (pLogS) to key toxicity endpoints like hERG inhibition and Hepatotoxicity. These models are maintained and continuously improved by Expert Systems, Inc., combining public and proprietary cheminformatics data for superior accuracy.
   
   
2. Transparency Through Uncertainty Quantification (UQ)
   In computational chemistry, knowing when not to trust a prediction is as important as the prediction itself. Every result from Q-Models includes Uncertainty Quantification (UQ) metrics. This transparency empowers scientists to assess model confidence and applicability domain, making Q-Models a reliable component of GxP-ready pipelines.
   
   
3. Proprietary Differentiators
   The partnership provides access to exclusive models that go beyond standard ADMET:

• Subcellular Localization (SubCell): Unique models predict the precise location of a compound across 15+ cellular compartments (e.g., Plasma Membrane, Lysosomes, Mitochondria). This is critical for drugs targeting specific organelles.
• LLM  — Powered Therapeutic Area Mapping: Leveraging advanced LLM models (such as Claude and GPT  — 4), Q-Models can map potential drug candidates to over 8,000 rare and common diseases, accelerating therapeutic area expansion and validation.

A True Partnership for the Future of R&D

This collaboration represents a significant step forward, combining Expert Systems' deep expertise in foundational model creation with Quantori's strengths in platform deployment, security, and enterprise-scale integration.

The introduction of Q-Models is set to dramatically decrease the time and cost associated with filtering and optimizing novel drug candidates, helping our clients bring life-saving therapeutics to patients faster.

To learn more about how to integrate Q-Models into your Q-Suite workflow, visit the Q-Suite website or reach out to us at contact@quantori.com.