Abstract

In the last four years, I have been exposed to various topics in scientific research under the supervision of Dr. Kraka in the CATCO group. Numerous involved chemistry projects were undertaken to gain an understanding of the basic laws of nature involving vibrational spectroscopy, molecular acidity, and catalysts based on transition metals for halogen chemistry. The insights from computational chemistry were then applied to model and predict various complicated problems in chemistry via artificial intelligence. With the help of classical artificial intelligence, the non-covalent interactions governing the properties of proteins and water properties were analyzed. Significant improvements were made in the field of drug discovery by the development of methods such as SSnet: to predict protein-ligand affinities, and CFGenNet: to generate novel drug candidates based on protein's history of interaction and towards solving a notorious quantum chemistry problem i.e. transition state prediction for small molecules by modeling TS-GAN: a generative model for guess transition state prediction. SSnet was further utilized to predict potential drugs for SARS-COV-2. Thus, it is evident that artificial intelligence holds strong potential in solving centuries-old complicated chemistry problems.

Degree Date

Summer 8-4-2021

Document Type

Dissertation

Degree Name

Ph.D.

Department

Chemistry

Advisor

Elfi Kraka

Subject Area

Biochemistry, Chemistry, Computer Science, Physics

Number of Pages

282

Format

.pdf

Creative Commons License

Creative Commons Attribution-Noncommercial 4.0 License
This work is licensed under a Creative Commons Attribution-Noncommercial 4.0 License

Available for download on Wednesday, July 27, 2022

Share

COinS