|
|
| Type |
Symposium |
| Area |
Data-Enabled Computational Chemistry |
| Room No. |
Room 324A |
| Time |
FRI 09:50-: |
| Code |
PHYS2-3 |
| Subject |
Deep learning for smart drug design |
| Authors |
Woo youn Kim
Department of Chemistry, Korea Advanced Institute of Science and Technology, Korea |
| Abstract |
The ultimate goal of materials and drug discovery is to create molecules with desired properties. This is obviously a difficult task because the molecular space is very large and discrete with a wide variety of molecules. For example, there are only 108 molecules synthesized, but 1060 molecules are estimated to be existing. Computer-aided molecular design is attracting attention as a promising solution for efficient materials and drug discovery. A fast calculation method allows us to find molecules with target properties through high-throughput virtual screening over known databases.
As an alternative strategy, we propose to use a molecular generative model based on machine learning for de novo molecular design. It is specialized in controlling multiple molecular properties simultaneously, embedding them in namely the latent space. As a proof of concept, we will show that it can be used to generate a number of molecules as drugs with specific properties. We also apply it to design new molecules with promising binding energy for a specific target protein and use them as potential drug candidates that are not in the database.
|
| E-mail |
wooyoun@kaist.ac.kr |
|
122nd General Meeting of the KCS