Kantify announces its new drug discovery technology

Kantify announces ZeptoNet, its new Virtual screening technology of small molecules

Why are new technologies, such as ZeptoNet, needed in drug discovery?

Drug discovery is known to be a lengthy and complex process. For this reason Kantify is working towards accelerating this procedure using its new virtual high throughput screening (vHTS) technology, ZeptoNet. Kantify has developed and trained ZeptoNet, a deep neural network, created for the purpose of predicting whether or not a drug candidate shows a certain type of bioactivity (e.g. inhibition or activation) on a protein.

The technology uses computer based (in silico) methods to perfect the early stages of drug discovery and, in this way, fasten the process. ZeptoNet therefore solves the core problem currently faced during these early stages of drug development, namely, the high amount of time that it takes to discover new drugs.

How does ZeptoNet work?

Kantify’s new technology has been trained on over 100 bioassays by making use of transfer learning. Furthermore, the technology has a novel featurization of both proteins and compounds. Finally, it operates active learning technologies with complex algorithms that can iteratively query a source of information in order to rapidly label new data points with the desired information.

How does ZeptoNet differentiate from previous technologies?

ZeptoNet’s key novelty is the fact that it is trained on multiple high throughput screening (HTS) campaign results. It therefore helps drug discovery efforts to move faster for new targets or existing targets. Instead of having to be trained from scratch - which is a common process in hit discovery efforts - ZeptoNet has predictive force for multiple targets, either new or existing. Kantify has succeeded in developing an AI model and supporting architecture which enables ZeptoNet to learn from many bioassays and their results. In this regard, Kantify’s approach is radically different from usual vHTS methods.

What are ZeptoNet's benefits?

ZeptoNet can solve a number of challenging problems faced in the drug discovery process such as the ones listed below.

  • Fast: ZeptoNet accelerates the finding of hits while having a superior performance.
  • Rich: ZeptoNet ‘s architecture enables to integrate a variety of data inputs. This leads to very accurate predictions, not only on the likeliness of a hit, but also on various physical and chemical properties.
  • Pretrained: ZeptoNet does not have to be trained from scratch as it is pretrained on several bioassays and already has an understanding of biophysical properties.
  • Data lean: As ZeptoNet is pretrained, it can start making accurate predictions even when there are very few known hits.

Find out more about ZeptoNet by reading Kantify’s white paper here