Technology

Artificial intelligence already touches every aspect of our lives, and has had a transformative impact across several of the world’s most competitive industries. It’s time to bring AI into drug discovery so that we can optimize the development of new medications for patients everywhere.

At InVivo AI, our mission is to reduce the time and cost involved in preclinical decision making, while also increasing the likelihood of success for compounds selected for clinical trials.

The Problem

It now takes 10-15 years and almost $2B to bring a new drug to market, largely due to th fact that ~90% of all compounds fail in clinical testing

40% of all these failures occur for toxicity reasons alone. Despite numerous preclinical studies, it remains nearly impossible to predict drug toxicity before running clinical trials, so much so that the FDA estimates that a 10% improvement in the ability to predict clinical trial failures in the early stages of drug development is worth almost $100M per approved drug.

Product Overview

Our proprietary deep learning platform integrates structural, target, and pathway-based descriptors to accurately generate the toxicological profiles of small molecule drugs in silico

Modules in Development

Clinical development

Clinical Development Module

Use our virtual screening platform to predict a compound’s likelihood of success in clinical testing, allowing you to rapidly filter for drug candidates likely to fail in clinical testing and to prioritize leads with the highest probabilities of eventual success.

Preclinical toxicology

Preclinical Toxicology Module

Use our in silico screening tools to simulate a broad selection of cell-based functional assays to better understand the toxicological profiles of your compounds early in the discovery process. Our meta-learning framework for predicting toxicity provides significantly greater predictive power and flexibility than existing models trained using similar datasets and architectures.

Drug-Drug interactions

Drug-Drug Interactions Module

Use our drug-drug interactions module to inform preclinical safety assessments by providing an exhaustive overview of potential adverse events if used in combination with existing medications.

Metabolite

Metabolite Module

Use our metabolite module to predict downstream toxicity risk arising from the processing of molecular metabolites. In combination with our other modules, this module can be used to inform structural modifications to reduce overall toxicity risk.

Explainer Functionality

By building explainability and interpretability into our platform, our aim is to allow for informed decision making to the greatest extent possible. Use our explainer toolkit to better understand which molecular fragments are driving the predicted outcome, as well as the mechanisms underlying any toxic responses by revealing relevant pathway components responsible for the classification.

Enhancer Functionality

Here we use deep generative models to drive lead optimization based upon the notion of molecular similarity. Our enhancer functionality can be used to propose molecular modifications to minimize adverse effects while having minimal impact on efficacy and/or binding at the target site.

Line
E 2

E2: Drive informed lead optimization by using the explainer module to identify negative critical fragments before optimizing compounds with the help of the enhancer module