SUPR
Deep learning for discovery of anti-infectives
Dnr:

NAISS 2025/5-311

Type:

NAISS Medium Compute

Principal Investigator:

Andreas Luttens

Affiliation:

Karolinska Institutet

Start Date:

2025-08-25

End Date:

2026-03-01

Primary Classification:

10407: Theoretical Chemistry

Secondary Classification:

10610: Bioinformatics and Computational Biology (Methods development to be 10203)

Tertiary Classification:

30103: Medicinal Chemistry (Natural Sciences at 10405)

Allocation

Abstract

This project aims to expand our computational drug discovery toolbox. To prepare ourselves against emerging infectious diseases, we must develop novel strategies that enable us to rapidly identify promising anti-infectives at minimal cost. Access to large collections of building blocks and robust organic synthesis procedures enabled chemical vendors to construct virtual databases containing several billions of readily synthesizable compounds. While this source of novel structures represents a major opportunity for drug discovery, synthesizing and experimentally evaluating the entire library remains intractable. Therefore, progress relies on computational tools capable of pinpointing which molecules to prioritize for synthesis and testing. Virtual screening is not restricted to compounds that are physically available, making it suitable to navigate virtual chemical spaces. One such technique is molecular docking, which predicts a molecule’s interaction with a target protein in seconds and enables evaluation of large databases. In this project, we will develop and apply such techniques to identify inhibitors targeting proteins essential for the survival or replication of pathogenic bacteria and viruses. All our projects are driven by molecular modeling and carried out in collaboration with experimentalists at other SciLifeLab research infrastructures.