AI Small Molecule Design

Posted by catalina on October 24th, 2023

Introduction of Small Molecules in Drug Discovery

Small molecules, a number of low molecular weight compounds, have been considered as perfect targets for designing new drugs in human disease therapy. The screened small molecules are tested in different types of cells, animal models, and eventually clinical trials. However, this normal drug discovery process can be time-consuming and least cost-effective. Recent studies have revealed that computer-based small molecule design tools have shown promising results in the process of small molecule design and optimization. Among them, AI-aided platform plays an important role in guiding the construction of small molecules that are of most relevance to targeted drugs.

For instance, deep learning (DL) methods are commonly used for achieving target recognition of small molecules in drug discovery. These DL models can successfully predict the activity and property of various small molecules. As one of the most popular DL architectures, a convolutional neural network (CNN) has been developed by analyzing the sparse connectivity and shared weights to mimic the interaction between candidate compounds and proteins. Besides, a big data library for training the DL models, including the AlphaGo, has been established to collect enough data for AI-aided intricate simulation.

Learn more: AI Small Molecule Design