News Release 

New artificial intelligence model to bridge biology and chemistry

Generative biology meets generative chemistry: Bidirectional conditional autoencoder to generate novel molecular structures for the desired transcriptional response

InSilico Medicine

IMAGE

IMAGE: New Artificial Intelligence Model to Bridge Biology and Chemistry view more 

Credit: Insilico Medicine

May 19th, 2020, Hong Kong - Insilico Medicine announces the publication of a new research paper titled "Molecular Generation for Desired Transcriptome Changes With Adversarial Autoencoders" in Frontiers in Pharmacology. This is the first study of this kind where novel molecular structures are created for a desired transcriptional response.

In this study, Insilico Medicine researchers developed a new model, the Bidirectional Adversarial Autoencoder, that learns a joint distribution of molecular structures and induced transcriptional response. The model can generate molecular structures for a given transcriptional response and vise versa. As a result, Insilico Medicine provided a model that combines both generative biology and generative chemistry. Using this model, researchers can run virtual screening, discover novel molecular structures, and predict transcriptional responses--one model to solve many problems.

"This paper shows that it is possible to generate novel molecular structures that induce the desired transcriptional response. At Insilico, we have been working on this project since 2016 and have created critical intellectual property covering the original ideas in generative biology proposed and patented by Alex Zhavoronkov and Alex Aliper. I hope that the generative chemistry and biology developed at Insilico will become household tools for big pharmaceutical companies. Many of these tools are available in our upcoming AI platform soon to be available for deployment at customer premises", said Daniil Polykovskiy, group leader at Insilico Medicine and senior author of the study.

###

About Insilico Medicine

Since 2014 Insilico Medicine is focusing on generative models, reinforcement learning (RL), and other modern machine learning techniques for the generation of new molecular structures with the specified parameters, generation of synthetic biological data, target identification, and prediction of clinical trials outcomes. Since its inception, Insilico Medicine raised over $52 million, published over 70 peer-reviewed papers, applied for over 20 patents, and received multiple industry awards.

Website http://insilico.com/

Media Contact

For further information, images or interviews, please contact:
ai@insilico.com

About Frontiers Research Topics

Frontiers' Research Topics are peer-reviewed article collections around themes of cutting-edge research. Defined, managed, and led by renowned researchers, they unite the world's leading experts around the hottest topics in research, stimulating collaboration and accelerating science.

About Frontiers in Pharmacology

Frontiers in Pharmacology is a leading journal in its field, publishing rigorously peer-reviewed research across disciplines, including basic and clinical pharmacology, medicinal chemistry, pharmacy and toxicology.

Disclaimer: AAAS and EurekAlert! are not responsible for the accuracy of news releases posted to EurekAlert! by contributing institutions or for the use of any information through the EurekAlert system.