Detecting similar binding pockets to enable systems polypharmacology

Duran‐Frigola, Miquel; Siragusa, Lydia; Ruppin, Eytan; Barril, Xavier; Cruciani, Gabriele; Aloy, Patrick

doi:10.1371/journal.pcbi.1005522

Cited by 40 publications

(31 citation statements)

References 69 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…If this is the case, such difference may be used to screen or design new more selective drug candidates. Similar approaches, using an ensemble of related ligand-binding site interactions or considering remote homologs of a target binding site, have been suggested previously 58 – 63 . We believe that the network of binding site communities will provide a basis for exploring more efficient computational approaches for drug discovery and design.…”

Section: Resultsmentioning

confidence: 88%

Global organization of a binding site network gives insight into evolution and structure-function relationships of proteins

Lee

Konc

Janežič

et al. 2017

Sci Rep

View full text Add to dashboard Cite

The global organization of protein binding sites is analyzed by constructing a weighted network of binding sites based on their structural similarities and detecting communities of structurally similar binding sites based on the minimum description length principle. The analysis reveals that there are two central binding site communities that play the roles of the network hubs of smaller peripheral communities. The sizes of communities follow a power-law distribution, which indicates that the binding sites included in larger communities may be older and have been evolutionary structural scaffolds of more recent ones. Structurally similar binding sites in the same community bind to diverse ligands promiscuously and they are also embedded in diverse domain structures. Understanding the general principles of binding site interplay will pave the way for improved drug design and protein design.

show abstract

Section: Resultsmentioning

confidence: 88%

Global organization of a binding site network gives insight into evolution and structure-function relationships of proteins

Lee

Konc

Janežič

et al. 2017

Sci Rep

View full text Add to dashboard Cite

show abstract

“…Protein Structure Comparison (PSC) is an essential task in structural biology and drug discovery; it allows researchers, for instance, to infer protein evolution (to understand better the relationship between protein structure and function) and to transfer knowledge about known proteins to a novel protein (Schenkel, Holm, Rosenström, & Kääriäinen, 2008). Some of the main applications of PSC include establishing structural, evolutionary, and functional relationships between proteins; assigning functional annotations to proteins (Mills, Beuning, & Ondrechen, 2015); drug repositioning (Haupt, Daminelli, & Schroeder, 2013); and identification of proteins with similar binding sites as potential targets for the same ligand (Duran-Frigola et al, 2017).…”

Section: Introductionmentioning

confidence: 99%

iProStruct2D: Identifying protein structural classes by deep learning via 2D representations

Nanni

Lumini

Pasquali

et al. 2020

Expert Systems with Applications

View full text Add to dashboard Cite

In this paper we address the problem of protein classification starting from a multi-view 2D representation of proteins. From each 3D protein structure, a large set of 2D projections is generated using the protein visualization software Jmol. This set of multi-view 2D representations includes 13 different types of protein visualizations that emphasize specific properties of protein structure (e.g., a backbone visualization that displays the backbone structure of the protein as a trace of the Cα atom). Each type of representation is used to train a different Convolutional Neural Network (CNN), and the fusion of these CNNs is shown to be able to exploit the diversity of different types of representations to improve classification performance. In addition, several multi-view projections are obtained by uniformly rotating the protein structure around its central X, Y, and Z viewing axes to produce 125 images. This approach can be considered a data augmentation method for improving the performance of the classifier and can be used in both the training and the testing phases. Experimental evaluation of the proposed approach on two datasets demonstrates the strength of the proposed method with respect to the other state-of-the-art approaches. The MATLAB code used in this paper is available at https://github.com/LorisNanni.

show abstract

“…In this analysis, we have focused exclusively on predictions available for all proteins of know sequences even for those that do not map to FunFams. Methods based on 3D structure are known to perform much better predictions, and only those can actually aspire to predict binding sites rather than binding residues [21][22][23][24][25].…”

Section: Discussionmentioning

confidence: 99%

FunFam protein families improve residue level molecular function prediction

Scheibenreif

Littmann

Orengo

et al. 2019

Preprint

View full text Add to dashboard Cite

Background The CATH database provides a hierarchical classification of protein domain structures including a sub-classification of superfamilies into functional families (FunFams). We analyzed the similarity of binding site annotations in these FunFams and incorporated FunFams into the prediction of protein binding residues. Results FunFam members agreed, on average, in 36.9±0.6% of their binding residue annotations. This constituted a 6.7-fold increase over randomly grouped proteins and a 1.2-fold increase (1.1-fold on the same dataset) over proteins with the same enzymatic function (identical Enzyme Commission, EC, number). Mapping de novo binding site prediction methods (BindPredict-CCS, BindPredict-CC) onto FunFam resulted in consensus predictions for those residues that were aligned and predicted alike (binding/non-binding) within a FunFam. This simple consensus increased the F1-score (for binding) 1.5-fold over the original prediction method. Variation of the threshold for how many proteins in the consensus prediction had to agree provided a convenient control of accuracy/precision and coverage/recall, e.g. reaching a precision as high as 60.8±0.4% for a stringent threshold. Conclusions The FunFams outperformed even the carefully curated EC numbers in terms of agreement of binding site residues. Additionally, we assume that our proof-of-principle through the prediction of protein binding residues will be relevant for many other solutions profiting from FunFams to infer functional information at the residue level.

show abstract

Detecting similar binding pockets to enable systems polypharmacology

Cited by 40 publications

References 69 publications

Global organization of a binding site network gives insight into evolution and structure-function relationships of proteins

Global organization of a binding site network gives insight into evolution and structure-function relationships of proteins

iProStruct2D: Identifying protein structural classes by deep learning via 2D representations

FunFam protein families improve residue level molecular function prediction

Contact Info

Product

Resources

About