SeRenDIP: SEquential REmasteriNg to DerIve profiles for fast and accurate predictions of PPI interface positions

Hou, Qingzhen; Geest, Paul F G De; Griffioen, Christian J; Abeln, Sanne; Heringa, Jaap; Feenstra, K. Anton

doi:10.1093/bioinformatics/btz428

Cited by 24 publications

(50 citation statements)

References 7 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…For clarity, we here summarize the training and testing protocols used to derive our random forest classifiers, following the procedure developed previously (Hou et al, 2017(Hou et al, , 2019, and also the details of constructing the Dset_anti dataset for sequence-based epitope prediction.…”

Section: Methodsmentioning

confidence: 99%

“…For each antigen sequence, we obtain a series of features per position across the query sequence, as previously described (Hou et al, 2017(Hou et al, , 2019. In short, for each input sequence, PSI-BLAST (version 2.2.22+; Altschul et al, 1997Altschul et al, , 2005Schäffer et al, 2001) is used to retrieve sequence homologs from the NR70 database using max.…”

Section: Generation Of Protein Sequence Featuresmentioning

confidence: 99%

“…We evaluate the performance of our trained models on the test sets, using the usual measures as used before (Hou et al, 2017(Hou et al, , 2019. Here, true positives (TP) are correct prediction of epitope residues, false positives (FP) are residues incorrectly predicted as epitope, true negatives (TN) designate non-interacting sites that were recognised, and false negatives (FN) are non-epitope sites which were predicted as epitope.…”

Section: Benchmarkingmentioning

confidence: 99%

“…This resulted in a widely usable interface predictor that only requires a single sequence as input (Hou et al, 2017(Hou et al, , 2019. It is based on a random forest model and incorporates several features that can be derived from the sequence, including conservation (Pirovano et al, 2006;Hou et al, 2015), predicted secondary structure (Guharoy and Chakrabarti, 2007;de Vries and Bonvin, 2008) and solvent accessible area (Ofran and Rost, 2007;Li et al, 2012), and as novel features protein length (Hou et al, 2015(Hou et al, , 2017 and predicted backbone flexibility (Cilia et al, 2013;Hou et al, 2017Hou et al, , 2019. Here we investigate if this approach using random forest models with previously selected features, and the combination of epitope data and general PPI data sets can help to make epitope prediction more accurate.…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

SeRenDIP-CE: Sequence-based Interface Prediction for Conformational Epitopes

Hou

Stringer

Waury

et al. 2020

Preprint

Self Cite

View full text Add to dashboard Cite

MotivationAntibodies play an important role in clinical research and biotechnology, with their specificity determined by the interaction with the antigen’s epitope region, as a special type of protein-protein interaction (PPI) interface. The ubiquitous availability of sequence data, allows us to predicting epitopes from sequence in order to focus time-consuming wet-lab experiments onto the most promising epitope regions. Here, we extend our previously developed sequence-based predictors for homodimer and heterodimer PPI interfaces to predict epitope residues that have the potential to bind an antibody.ResultsWe collected and curated a high quality epitope dataset from the SAbDaB database. Our generic PPI heterodimer predictor obtained an AUC-ROC of 0.666 when evaluated on the epitope test set. We then trained a random forest model specifically on the epitope dataset, reaching AUC 0.694. Further training on the combined heterodimer and epitope datasets, improves our final predictor to AUC 0.703 on the epitope test set. This is better than the best state-of-the-art sequence-based epitope predictor BepiPred-2.0. On one solved antibody-antigen structure of the COVID19 virus spike RNA binding domain, our predictor reaches AUC 0.778. We added the SeRenDIP-CE Conformational Epitope predictors to our webserver, which is simple to use and only requires a single antigen sequence as input, which will help make the method immediately applicable in a wide range of biomedical and biomolecular research.AvailabilityWebserver, source code and datasets are available at www.ibi.vu.nl/programs/serendipwww/Contactk.a.feenstra@vu.nl

show abstract

Section: Methodsmentioning

confidence: 99%

Section: Generation Of Protein Sequence Featuresmentioning

confidence: 99%

Section: Benchmarkingmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

SeRenDIP-CE: Sequence-based Interface Prediction for Conformational Epitopes

Hou

Stringer

Waury

et al. 2020

Preprint

Self Cite

View full text Add to dashboard Cite

show abstract

“…On each target sequence, the backbone dynamics (DynaMine) 1,2 , and related side-chain dynamics and conformational propensities 3 were predicted at the per-amino acid level, as well as early folding (EFoldMine) 3 , disorder (DisoMine) 4 , b-sheet aggregation (Agmata) 5 , proteinprotein interactions (SeRenDIP) 6,10 , and SeRenDIP-CE epitope propensities 7 . Predictions of FUS-like phase separation were also performed with PSPer 11 .…”

Section: Predictionsmentioning

confidence: 99%

Online biophysical predictions for SARS-CoV-2 proteins

Kagami¹,

Roca-Martínez

Gavaldá-García

et al. 2020

Preprint

View full text Add to dashboard Cite

The SARS-CoV-2 virus, the causative agent of COVID-19, consists of an assembly of proteins that determine its infectious and immunological behavior, as well as its response to therapeutics. Major structural biology efforts on these proteins have already provided essential insights into the mode of action of the virus, as well as avenues for structure-based drug design. However, not all of the SARS-CoV-2 proteins, or regions thereof, have a well-defined three-dimensional structure, and as such might exhibit ambiguous, dynamic behaviour that is not evident from static structure representations, nor from molecular dynamics simulations using these structures. We here present a website (http://sars2.bio2byte.be/) that provides protein sequence-based predictions of the backbone and side-chain dynamics and conformational propensities of these proteins, as well as derived early folding, disorder, β-sheet aggregation and protein-protein interaction propensities. These predictions attempt to capture the ‘emergent’ properties of the proteins, so the inherent biophysical propensities encoded in the sequence, rather than context-dependent behaviour such as the final folded state. In addition, we provide an indication of the biophysical variation that is observed in homologous proteins, which give an indication of the limits of the functionally relevant biophysical behaviour of these proteins. With this website, we therefore hope to provide researchers with further clues on the behaviour of SARS-CoV-2 proteins.

show abstract

Machine learning solutions for predicting protein–protein interactions

Casadio

Martelli

Savojardo

2022

WIREs Comput Mol Sci

View full text Add to dashboard Cite

Proteins are "social molecules." Recent experimental evidence supports the notion that large protein aggregates, known as biomolecular condensates, affect structurally and functionally many biological processes. Condensate formation may be permanent and/or time dependent, suggesting that biological processes can occur locally, depending on the cell needs. The question then arises as to which extent we can monitor protein-aggregate formation, both experimentally and theoretically and then predict/simulate functional aggregate formation. Available data are relative to mesoscopic interacting networks at a proteome level, to protein-binding affinity data, and to interacting protein complexes, solved with atomic resolution. Powerful algorithms based on machine learning (ML) can extract information from data sets and infer properties of never-seen-before examples. ML tools address the problem of proteinprotein interactions (PPIs) adopting different data sets, input features, and architectures. According to recent publications, deep learning is the most successful method. However, in ML-computational biology, convincing evidence of a success story comes out by performing general benchmarks on blind data sets. Results indicate that the state-of-the-art ML approaches, based on traditional and/or deep learning, can still be ameliorated, irrespectively of the power of the method and richness in input features. This being the case, it is quite evident that powerful methods still are not trained on the whole possible spectrum of PPIs and that more investigations are necessary to complete our knowledge of PPI-functional interactions.

show abstract

SeRenDIP: SEquential REmasteriNg to DerIve profiles for fast and accurate predictions of PPI interface positions

Cited by 24 publications

References 7 publications

SeRenDIP-CE: Sequence-based Interface Prediction for Conformational Epitopes

SeRenDIP-CE: Sequence-based Interface Prediction for Conformational Epitopes

Online biophysical predictions for SARS-CoV-2 proteins

Machine learning solutions for predicting protein–protein interactions

Contact Info

Product

Resources

About