Authors: Professor Anita Mahajan, Ajaz Shaikh, Shubham Ghume, Neeraj Lonkar, Saif Shaikh
Abstract: Drug development research is traditionally a lengthy, resource-intensive, and expensive process, often relying on experimental approaches and iterative laboratory trials. the emergence of generative adversarial networks (gans) hasintroduced a novel and efficient approach to this field by facilitating the generation of new molecular structures. This research explores the application of molgan, a specialized gan framework tailored for generating molecular graphs in drug discovery. traditional methods struggle with inefficiencies and the vastness of the chemical space, making it challenging to identify molecules with specific pharmacological properties. molgan addresses these limitations by automating molecular generation while incorporating desired chemical characteristics. by leveraging reinforcement learning techniques, molgan fine- tunes the generation process to produce drug-like molecules, enhancing both the speed and effectiveness of drug discovery efforts.
DOI: http://doi.org/
