“The Covid-19 pandemic shone a light on the pitfalls of the traditional NCE [new chemical entity] process that takes many years, costs a lot of money, and has a massive failure rate” – Andrew Watson, VP of Artificial Intelligence at Cambridge-based drug discovery firm Healx
- One aspect of AI that is still problematic is neural networks and their lack of transparency for health applications. To trust a decision, we need to know why that decision was made.
- Designing a novel molecule against a new target is inherently a small data problem – there is simply less accumulated data available for newly identified targets.
Seongok Ryu, Jaechang Lim, Sang-Yeon Hwang, Jeong-eun Park
Wonho Zhung, Seokhyun Moon, and Woo Youn Kim
Drug-target interaction (DTI)
Predicting a binding affinity of a compound against the target protein, a.k.a. drug-target interaction (DTI), is one of the most essential properties in a drug discovery process.
Here, we utilize the 3D structure of protein-ligand complexes to predict the drug-target interaction by predicting atom-atom pairwise energy with physics-informed parameterized equations.
Scoring the binding affinity of drug candidates against the target with our DTI model enables the high-throughput virtual screening scenario for finding novel compounds.
Molecular generative model for drug design
The molecular generative model samples highly diverse molecules inferred from the targeted chemical space.
Thus, the generative model can increase the probability of finding the hit compound with respect to the original chemical space.
Various fields in drug design such as the hit-to-lead process can take an advantage of the generative model by optimize target properties and preserve desirable characteristics.