The Meta-QSAR Project - How to Design Drugs

The Problem: There is no single best way of learning and applying QSARs (Quantitative Structure Activity Relationship), nor could such a method exist. Instead it is clear from theory and practice that some target-type/ compound-type/ molecular-representation/ learning-method/ approaches work better together than others. However, despite the vast size of the QSAR literature previous comparative studies have only compared a limited number of QSAR problem combinations. Therefore, currently the QSAR scientist has little to guide her/him on which QSAR approach to choose for a specific problem.

The Aim: The aim of the Meta-QSAR project is to utilise newly available public domain chemoinformatic databases and in-house datasets to systematically run extensive comparative QSAR experiments. We will then generalise these results to learn which target-type/ compound-type/ compound-representation /learning-method combinations work best together. We will learn how to better apply existing QSAR methods. This is meta-learning, using machine learning to learn about QSAR leaning. We will make the knowledge we learn publically available to guide future QSAR learning. This will decrease the time and cost to develop new drugs.

For inquiries, please contact Ivan Olier