P2Rank: machine learning based tool for rapid and accurate prediction of ligand binding sites from protein structure

@article{Krivk2018P2RankML,
  title={P2Rank: machine learning based tool for rapid and accurate prediction of ligand binding sites from protein structure},
  author={Radoslav Kriv{\'a}k and David Hoksza},
  journal={Journal of Cheminformatics},
  year={2018},
  volume={10}
}
AbstractBackgroundLigand binding site prediction from protein structure has many applications related to elucidation of protein function and structure based drug discovery. It often represents only one step of many in complex computational drug design efforts. Although many methods have been published to date, only few of them are suitable for use in automated pipelines or for processing large datasets. These use cases require stability and speed, which disqualifies many of the recently… 

Exploring the computational methods for protein-ligand binding site prediction

PocketAnchor: Learning Structure-Based Pocket Representations for Protein-Ligand Interaction Prediction

TLDR
This work proposes a novel structure-based protein representation method, named PocketAnchor, for capturing the local environmental and spatial features of protein pockets to facilitate protein-ligand interaction-related learning tasks and exhibits great generalization ability for novel proteins.

3DLigandSite: structure-based prediction of protein–ligand binding sites

TLDR
A machine learning element is introduced as the final prediction step of 3DLigandSite, which improves the accuracy of predictions and provides a confidence score for each residue predicted to be part of a binding site.

Computational methods and tools for binding site recognition between proteins and small molecules: from classical geometrical approaches to modern machine learning strategies

TLDR
This review describes the most recent developments in protein function recognition and binding site prediction, in terms of both freely-available and commercial solutions and tools, detailing the main characteristics of the considered tools and providing a comparative analysis of their performance.

PickPocket: Pocket binding prediction for specific ligand families using neural networks

TLDR
PickPocket tool, a tool that focuses on a small set of user-defined ligands and uses neural networks to train a ligand-binding prediction model, thinks that the PickPocket tool can help to discover new protein functions by investigating the binding sites of specific ligand families.

PrankWeb: web server for ligand binding-site prediction and visualization

TLDR
PrankWeb is an online resource providing an interface to P2Rank, a state-of-the-art ligand binding site prediction method which is based on the prediction of ligandability of local chemical neighborhoods centered on points placed on a solvent accessible surface of a protein.

GRaSP-web: a machine learning strategy to predict binding sites based on residue neighborhood graphs

TLDR
A web server that uses GRaSP (Graph-based Residue neighborhood Strategy to Predict binding sites), a residue-centric method based on graphs that uses machine learning to predict putative ligand binding site residues, which proved to be consistent in predicting binding sites for bound/unbound structures.

DeepBindPoc: a deep learning method to rank ligand binding pockets using molecular vector representation

TLDR
The systematic testing and validation of the proposed Deep BindPoc method suggest that DeepBindPoc is a valuable tool to rank near-native pockets for theoretically modeled protein with unknown experimental active site but have known ligand.

PrankWeb: a web server for ligand binding site prediction and visualization

TLDR
PrankWeb is an online resource providing an interface to P2Rank, a state-of-the-art method for ligand binding site prediction based on the prediction of local chemical neighborhood ligandability centered on points placed on a solvent-accessible protein surface.

Sequence-based prediction of protein binding regions and drug–target interactions

TLDR
A deep learning model is constructed, named Highlights on Target Sequences (HoTS), which predicts binding regions (BRs) between a protein sequence and a drug ligand, as well as DTIs between them, which showed good performance in BR prediction on independent test datasets even though it does not use 3D structure information in its prediction.
...

References

SHOWING 1-10 OF 87 REFERENCES

eFindSite: Improved prediction of ligand binding sites in protein models using meta-threading, machine learning and auxiliary ligands

TLDR
Carefully calibrated confidence estimates strongly indicate that highly reliable ligand binding predictions are made for the majority of gene products, thus eFindSite holds a significant promise for large-scale genome annotation and drug development projects.

P2RANK: Knowledge-Based Ligand Binding Site Prediction Using Aggregated Local Features

TLDR
P2RANK is introduced, a novel machine learning-based method for prediction of ligand binding sites from protein structure that outperforms Fpocket on three representative datasets and represents an ideal tool for inclusion into future structural bioinformatics pipelines.

Identification of cavities on protein surface using multiple computational approaches for drug binding site prediction

TLDR
This work adds four more methods Fpocket, GHECOM, ConCavity and POCASA to further improve the prediction success rate of MetaPocket, and constructs a non-redundant dataset of drug-target complexes with known structure from DrugBank, DrugPort and PDB database to predict drug binding sites.

bSiteFinder, an improved protein-binding sites prediction server based on structural alignment: more accurate and less time-consuming

TLDR
bSiteFinder as a protein-binding sites prediction server was developed based on the largest database of bound templates so far with stringent quality control, and the efficiency of the server has been significantly improved by introducing Homology Indexing, Chain length indexing, Stability of Complex and Optimized Multiple-Templates Clustering.

A threading-based method (FINDSITE) for ligand-binding site prediction and functional annotation

TLDR
FINDSITE is described, a method for ligand-binding site prediction and functional annotation based on binding-site similarity across groups of weakly homologous template structures identified from threading that can accurately assign a molecular function to the protein model.

DeepSite: protein‐binding site predictor using 3D‐convolutional neural networks

TLDR
This work presents a novel knowledge‐based approach that uses state‐of‐the‐art convolutional neural networks, where the algorithm is learned by examples, and demonstrates superior performance to two other competitive algorithmic strategies.

Predicting Ligand Binding Sites on Protein Surfaces by 3-Dimensional Probability Density Distributions of Interacting Atoms

TLDR
The structure-based machine learning algorithm ISMBLab-LIG was developed to predict LBSs on protein surfaces with input attributes derived from the three-dimensional probability density maps of interacting atoms, which were reconstructed on the query protein surfaces and were relatively insensitive to local conformational variations of the tentative ligand binding sites.

LIBRUS: combined machine learning and homology information for sequence-based ligand-binding residue prediction

TLDR
A sequence-based method that combines homology-based transfer and direct prediction using machine learning and is slightly more discriminating than a support vector machine learner using profiles and predicted secondary structure is developed.

Comparison and Druggability Prediction of Protein-Ligand Binding Sites from Pharmacophore-Annotated Cavity Shapes

TLDR
A novel binding site description (VolSite) coupled to an alignment and comparison tool (Shaper) combining the speed of alignment-free methods with the interpretability of aligning sites based on the comparison of negative images of binding cavities.
...