SUPR
Faster protein-protein interaction prediction
Dnr:

NAISS 2025/5-21

Type:

NAISS Medium Compute

Principal Investigator:

Lionel Guy

Affiliation:

Uppsala universitet

Start Date:

2025-01-30

End Date:

2025-09-01

Primary Classification:

10601: Structural Biology

Webpage:

Allocation

Abstract

Proteins play a crucial role in all essential cellular functions, with their tertiary structure being vital to understanding their function. Traditional experimental methods, such as X-ray crystallography or Cryo-EM, are time-consuming and labor-intensive for determining protein structures. Thus, researchers have long sought to develop in silico alternatives. In recent decades, advancements in protein structure prediction have accelerated, with AlphaFold3 marking a breakthrough in the field. However, limited accessibility has hindered its widespread adoption by the research community. Recently, we have developed a novel approach that allows us to predict protein structure up to three-fold faster compared to AlphaFold3 while maintaining similar prediction accuracy. Moreover, this can be useful for faster prediction of protein-protein interaction on a large scale.