Project: Mapping every binding site to design more selective drugs

Challenge Space

Health - Resistance

Prinicipal Investigator(s) and team names

Dr Darren Fayne, Dr Ant Chubb, with Muzammil Kabier and Alex Pegeeva, Molecular Design Group

Collaborating Institutions

Funding support

Project start date

2026

End date

Current

Problem

A great many drug candidates fail late, and often because they bind something they were never meant to. Selectivity is normally tested one protein at a time, near the end of a project, when a decade of work is already committed. 

The same gap limits anti-infective discovery: without a way to tell which protein pockets in a bacterium or parasite have no counterpart in human cells, target choice stays narrow and conservative.

Background

The group holds a Research Ireland Frontiers for the Future award to build the Human Pocketome Pharmacophore, a systematic description of every place in the human proteome where a small molecule could bind, recorded not as a picture but as a signature of the chemistry each site demands. 

Advances in protein structure prediction mean a catalogue of this scale is achievable for the first time. Both leads have worked at whole-proteome scale before, including tools that identify candidate targets and leads directly from genomic data.

Solution

The atlas is being assembled and will be released so that any project can ask, early and cheaply, what else in the body resembles the site it is aiming at. 

For resistance research the more interesting question is the reverse one. Running the same analysis over the proteome of a pathogen and comparing against the human atlas identifies pockets that exist in the pathogen and nowhere in us, which is a principled definition of a selective target and a route to widening the very narrow set of proteins anti-infective discovery currently relies on. Extending the method in that direction is a planned strand of the group's work.