SHREC 2022: Protein-ligand binding site recognition

TL;DR

This paper reviews methods from the SHREC 2022 contest on protein-ligand binding site recognition, highlighting the challenges and competitiveness of simple approaches versus complex data-driven algorithms in identifying binding pockets.

Contribution

It provides an analysis of various computational methods for binding site recognition, emphasizing the persistent challenges and the surprising competitiveness of non-machine-learning techniques.

Findings

Simple non-machine-learning methods are highly competitive.

Binding pocket detection remains a challenging learning problem.

Data imbalance affects the accuracy of pocket recognition algorithms.

Abstract

This paper presents the methods that have participated in the SHREC 2022 contest on protein-ligand binding site recognition. The prediction of protein-ligand binding regions is an active research domain in computational biophysics and structural biology and plays a relevant role for molecular docking and drug design. The goal of the contest is to assess the effectiveness of computational methods in recognizing ligand binding sites in a protein based on its geometrical structure. Performances of the segmentation algorithms are analyzed according to two evaluation scores describing the capacity of a putative pocket to contact a ligand and to pinpoint the correct binding region. Despite some methods perform remarkably, we show that simple non-machine-learning approaches remain very competitive against data-driven algorithms. In general, the task of pocket detection remains a challenging learning problem which suffers of intrinsic difficulties due to the lack of negative examples (data imbalance problem).