Protein-ligand binding site recognition using complementary binding-specific substructure comparison and sequence profile alignment

By Jianyi Yang, Ambrish Roy, Yang Zhang. 2013

942 citations recorded by the source corpus.

MOTIVATION Identification of protein-ligand binding sites is critical to protein function annotation and drug discovery. However, there is no method that could generate optimal binding site prediction for different protein types. Combination of complementary predictions is probably the most reliable solution to the problem. RESULTS We develop two new methods, one based on binding-specific substructure comparison (TM-SITE) and another on sequence profile alignment (S-SITE), for complementary binding site predictions. The methods are tested on a set of 500 non-redundant proteins harboring 814 natural, drug-like and metal ion molecules. Starting from low-resolution protein structure predictions, the methods successfully recognize >51% of binding residues with average Matthews correlation coefficient (MCC) significantly higher (with P-value <10(-9) in student t-test) than other state-of-the-art methods, including COFACTOR, FINDSITE and ConCavity. When combining TM-SITE and S-SITE with other structure-based programs, a consensus approach (COACH) can increase MCC by 15% over the best individual predictions. COACH was examined in the recent community-wide COMEO experiment and consistently ranked as the best method in last 22 individual datasets with the Area Under the Curve score 22.5% higher than the second best method. These data demonstrate a new robust approach to protein-ligand binding site recognition, which is ready for genome-wide structure-based function annotations. AVAILABILITY http://zhanglab.ccmb.med.umich.edu/COACH/

Subfields: Computational Theory and Mathematics, Molecular Biology

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computational Theory and Mathematics

In Molecular Biology

Citations in the corpus

I-TASSER server: new development for protein structure and function predictions (2015, methodology): “Starting from the computational structure modeling, a number of methods have been proposed to annotate the biological function of protein molecules, with examples including COFACTROR (8,9), COACH (10), ConCavity (11), FINDSITE (12), Firestar (13) and 3DLigandSite (14), among many others.”
Geometric Transformers for Protein Interface Contact Prediction (2021, background): “This threshold of 30% sequence identity is commonly used in the bioinformatics literature (Jordan et al. (2012), Yang et al. (2013)) to prevent large evolutionary overlap between a dataset’s cross-validation partitions.”
PrankWeb: a web server for ligand binding site prediction and visualization (2019, methodology): “Prediction speeds varied greatly between tools, ranging from under one second (Fpocket, P2Rank) to > 10 h (COACH) for prediction on one average sized protein (2500 atoms).”
COACH-D: improved protein–ligand binding sites prediction with refined ligand-binding poses through molecular docking (2018, positive): “The detailed descriptions of these methods are available in the original COACH algorithm (7) and the publications of the corresponding methods (4–6).”
PointSite: a point cloud segmentation tool for identification of protein ligand binding atoms (2019, positive): “For all other datasets, our approach is 7%, 11%, 14%, 14%, and 15% better than P2Rank on DT198, B277, ASTEX85, COACH420, and HOLO4k, respectively.”
A chronotherapeutics-applicable multi-target therapeutics based on AI: Example of therapeutic hypothermia (2022, methodology): “Four of these methods are COFAC-TOR [27], FINDSITE [28], TMSITE [29] and S-SITE [29].”
CB-Dock: a web server for cavity detection-guided protein–ligand blind docking (2019, methodology): “We compared our method (called CurPocket) with state-of-the-art protein – ligand binding site prediction methods using the benchmark set of COACH [23], which is one of the best prediction methods.”
Protein Structure and Function Prediction Using I-TASSER (2015, methodology): “I-TASSER-based algorithms were extensively tested in both benchmark studies (Wu et al., 2007; Zhang et al., 2011; Roy et al., 2012; Yang et al., 2013b) and blind tests (Zhang, 2007; Zhang, 2009; Xu et al., 2011; Zhang, 2014).”
Computational methods and tools for binding site recognition between proteins and small molecules: from classical geometrical approaches to modern machine learning strategies (2019, methodology): “edu/COACH / Consensus method Web/Desktop COACH benchmark dataset [57]”
End-to-End Learning on 3D Protein Structure for Interface Prediction (2018, methodology): “This is a commonly used sequence identity threshold [33], [25], but competing methods for protein interface prediction do not employ such pruning on their training set, which may bias performance comparisons in their favor.”