Computational Prediction of Disordered Protein Motifs Using SLiMSuite

Edwards RJ, Paulsen K, Aguilar Gomez CM & Pérez-Bercoff Å (2020): Computational Prediction of Disordered Protein Motifs using SLiMSuite. Methods Mol Biol. 2141:37-72. doi: 10.1007/978-1-0716-0524-0_3. [PubMed]

Abstract

Short linear motifs (SLiMs) are important mediators of interactions between intrinsically disordered regions of proteins and their interaction partners. Here, we detail instructions for the computational prediction of SLiMs in disordered protein regions, using the main tools of the SLiMSuite package: (1) SLiMProb identifies and calculates enrichment of predefined motifs in a set of proteins; (2) SLiMFinder predicts SLiMs de novo in a set of proteins, accounting for evolutionary relationships; (3) QSLiMFinder increases SLiMFinder sensitivity by focusing SLiM prediction on a specific query protein/region; (4) CompariMotif compares predicted SLiMs to known SLiMs or other SLiM predictions to identify common patterns. For each tool, command-line and online server examples are provided. Detailed notes provide additional advice on different applications of SLiMSuite, including batch running of multiple datasets and conservation masking using alignments of predicted orthologues.

#ABACBS2015 Poster P20: SLiMScape 3.x: a Cytoscape 3 app for discovery of Short Linear Motifs in protein interaction networks

Emily Olorin, Kevin T. O’Brien, Nicolas Palopoli, Åsa Pérez-Bercoff, Denis C. Shields & Richard J. Edwards. http://f1000research.com/posters/4-1024.

Abstract

Short linear motifs (SLiMs) are small protein sequence patterns that mediate a large number of critical protein-protein interactions, involved in processes such as complex formation, signal transduction, localisation and stabilisation. SLiMs show rapid evolutionary dynamics and are frequently the targets of molecular mimicry by pathogens. Identifying enriched sequence patterns due to convergent evolution in non-homologous proteins has proven to be a successful strategy for computational SLiM prediction. Tools of the SLiMSuite package use this strategy, using a statistical model to identify SLiM enrichment based on the evolutionary relationships, amino acid composition and predicted disorder of the input proteins. The quality of input data is critical for successful SLiM prediction. Cytoscape provides a user-friendly, interactive environment to explore interaction networks and select proteins based on common features, such as shared interaction partners. SLiMScape embeds tools of the SLiMSuite package for de novo SLiM discovery (SLiMFinder and QSLiMFinder) and identifying occurrences/enrichment of known SLiMs (SLiMProb) within this interactive framework. SLiMScape makes it easier to (1) generate high quality hypothesis-driven datasets for these tools, and (2) visualise predicted SLiM occurrences within the context of the network. To generate new predictions, users can select nodes from a protein network or provide a set of Uniprot identifiers. SLiMProb also requires additional query motif input. Jobs are then run remotely on the SLiMSuite server (http://rest.slimsuite.unsw.edu.au) for subsequent retrieval and visualisation. SLiMScape can also be used to retrieve and visualise results from jobs run directly on the server.

SLiMScape 3.x: a Cytoscape 3 app for discovery of Short Linear Motifs in protein interaction networks

The latest paper from the lab (featuring work from Emily Olirin’s summer project) is now out at F1000Research. We’ve not gone down the post-publication peer review route before, so it will be interesting to see how that goes, but it made sense in this case as it’s part of a special Cytoscape Apps channel. Rest assured that it has undergone technical review, just not scientific review. [At time of submission: peer review now complete.]

Olorin E, O’Brien KT, Palopoli N, Pérez-Bercoff A, Shields DC, Edwards RJ (2015): SLiMScape 3.x: a Cytoscape 3 app for discovery of Short Linear Motifs in protein interaction networks [version 1; referees: 2 approved]. F1000Research 4:477. (doi: 10.12688/f1000research.6773.1)

Abstract

Short linear motifs (SLiMs) are small protein sequence patterns that mediate a large number of critical protein-protein interactions, involved in processes such as complex formation, signal transduction, localisation and stabilisation. SLiMs show rapid evolutionary dynamics and are frequently the targets of molecular mimicry by pathogens. Identifying enriched sequence patterns due to convergent evolution in non-homologous proteins has proven to be a successful strategy for computational SLiM prediction. Tools of the SLiMSuite package use this strategy, using a statistical model to identify SLiM enrichment based on the evolutionary relationships, amino acid composition and predicted disorder of the input proteins. The quality of input data is critical for successful SLiM prediction. Cytoscape provides a user-friendly, interactive environment to explore interaction networks and select proteins based on common features, such as shared interaction partners. SLiMScape embeds tools of the SLiMSuite package for de novo SLiM discovery (SLiMFinder and QSLiMFinder) and identifying occurrences/enrichment of known SLiMs (SLiMProb) within this interactive framework. SLiMScape makes it easier to (1) generate high quality hypothesis-driven datasets for these tools, and (2) visualise predicted SLiM occurrences within the context of the network. To generate new predictions, users can select nodes from a protein network or provide a set of Uniprot identifiers. SLiMProb also requires additional query motif input. Jobs are then run remotely on the SLiMSuite server (http://rest.slimsuite.unsw.edu.au) for subsequent retrieval and visualisation. SLiMScape can also be used to retrieve and visualise results from jobs run directly on the server. SLiMScape and SLiMSuite are open source and freely available via GitHub under GNU licenses.

QSLiMFinder: improved short linear motif prediction using specific query protein data

Palopoli N, Lythgow KT & Edwards RJ (2015). QSLiMFinder: improved short linear motif prediction using specific query protein data. Bioinformatics 31(14): 2284-2293. [PDF]

Abstract

Motivation: The sensitivity of de novo short linear motif (SLiM) prediction is limited by the number of patterns (the motif space) being assessed for enrichment. QSLiMFinder uses specific query protein information to restrict the motif space and thereby increase the sensitivity and specificity of predictions.

Results: QSLiMFinder was extensively benchmarked using known SLiM-containing proteins and simulated protein interaction datasets of real human proteins. Exploiting prior knowledge of a query protein likely to be involved in a SLiM-mediated interaction increased the proportion of true positives correctly returned and reduced the proportion of datasets returning a false positive prediction. The biggest improvement was seen if a short region of the query protein flanking the interaction site was known.

Availability and Implementation: All the tools and data used in this study, including QSLiMFinder and the SLiMBench benchmarking software, are freely available under a GNU license as part of SLiMSuite, at: http://bioware.soton.ac.uk.

Supplementary information: Supplementary data are available at the journal’s web site and at http://bioware.soton.ac.uk/research/qslimfinder/.

PMID: 25792551

ABiC2014 posters on F1000Posters

Better late than never, both posters from ABiC2014 are now on F1000Posters, along with many other great posters from the conference:

Richard J Edwards, Ranjeeta Menon, Nicolas Palopoli & Jason FH Wong (2015) Molecular mimicry in viruses and cancer. F1000Posters 6: 101.

Nicolas Palopoli & Richard J Edwards (2014) Computational prediction of protein interaction motifs using interaction networks and 3-dimensional structures. F1000Posters 5: 1682.

#ABiC14 Poster 46: Computational prediction of protein interaction motifs from integrated protein sequence, structure and interaction data

Nicolas Palopoli & Richard J Edwards

Abstract

Protein-protein interactions (PPI) between globular domains and Short Linear Motifs (SLiMs) play a crucial part in many biological processes. SLiMs are short stretches of 5 to 15 amino acids with high evolutionary plasticity that are usually found in disordered regions of proteins. Their role as ligands for molecular signalling, post-translational modifications and subcellular targeting has been increasingly studied over recent years, but experimental discovery of SLiMs remains a challenging task due to their small size and high degeneracy. As a consequence, computational tools for prediction and analysis of SLiMs are a valuable resource.

We have previously developed SLiMFinder, a motif discovery tool that applies a model of convergent evolution to estimate the statistical significance of over-represented motifs with high specificity. Here, we aim to improve motif discovery by integrating SLiMFinder with methods that predict new domain-motif interactions directly from structural features in high-resolution 3D data. To this end we have developed “Query” SLiMFinder (QSLiMFinder), which uses knowledge of the interaction interface to constrain the motif search space and thereby increase search sensitivity. We have benchmarked QSLiMFinder using the Eukaryotic Linear Motif (ELM) database and simulated data. As expected, specific domain-motif interaction data can increase the power of de novo SLiM prediction from a set of proteins with a common PPI partner.

We are now applying QSLiMFinder to large-scale analysis of public PPI and 3D structure data. Domain-motif interactions are predicted from structures in the protein data bank (PDB). QSLiMFinder then identifies patterns within the putative motif region that are over-represented in the other known PPI partners of the domain-containing protein. This will add crucial molecular details to the interactome.

InCoB 2014: Computational prediction of molecular mimicry in host-pathogen protein-protein interactions

Ranjeeta Menon, Nicolas Palopoli & Richard J. Edwards. F1000Research 2014, 5:1027 (poster)

Abstract

Many viruses hijack important cellular machinery of their hosts through the convergent evolution of host short linear motifs (SLiMs). We are combining host-pathogen and host-host protein-protein interaction data to explore such “molecular mimicry” across a broad range of human viruses. Computational tools from the SLiMSuite package are being used to assess the potential for bioinformatic analysis of high-throughput experiments to identify novel cases of molecular mimicry.

PPI-Net Poster: Computational prediction of short linear motifs integrating protein-protein interactions, sequence and structural data

Nico’s poster from the 3rd Protein-Protein Interaction Network (PPI-Net) Young Researchers Symposium is now available on F1000Posters:

Nicolas Palopoli & Richard J Edwards (2014) Computational prediction of short linear motifs integrating protein-protein interactions, sequence and structural data. F1000Posters 5: 407.

Abstract

Protein-protein interactions (PPI) between globular domains and Short Linear Motifs (SLiMs) play a crucial part in many biological processes. SLiMs are short stretches of 5 to 15 amino acids with high evolutionary plasticity that are usually found in disordered regions of proteins. Their role as ligands for molecular signalling, post-translational modifications and subcellular targeting has been increasingly studied over recent years, but experimental discovery of SLiMs remains a challenging task due to their small size and high degeneracy. As a consequence, computational tools for prediction and analysis of SLiMs are a valuable resource.

We have previously presented SLiMFinder, a motif discovery tool that applies a model of convergent evolution to estimate the statistical significance of over-represented motifs. In this project we aim to improve motif discovery by integrating SLiMFinder with methods that predict new domain-motif interactions directly from structural features in high-resolution 3D data. To this end we have developed “Query” SLiMFinder (QSLiMFinder) which uses knowledge of the interaction interface to constrain the motif search space and thereby increase search sensitivity. Using putative SLiM-carrying regions extracted from protein structures as queries and targeting data from public protein-protein interaction databases, we applied QSLiMFinder to find over-represented recurring sequence patterns from proteins that all share a common interaction partner.

Benchmarking of QSLiMFinder shows that specific domain-motif interaction data help in finding novel instances of known motifs or entire de novo SLiMs, improving over results returned by PPI data alone. SLiM discovery capitalizes from the availability of experimentally identified PPIs with high-quality predictions of the interacting sites. The methods developed here help to enhance annotation in public databases of SLiMs and could be used to mine new PPI data as it becomes available, adding molecular detail to interactome networks.

Nicolas Palopoli (Postdoctoral Research Fellow)

Nico Palopoli started training in Computational Biology with his undergraduate thesis project on the structural modeling and characterization of starch-synthase III from Arabidopsis thaliana, under the guidance of Dr. Gustavo Parisi at the Structural Bioinformatics Group from Universidad Nacional de Quilmes (Argentina). He stayed at the group to fulfill his PhD thesis on the validation of protein 3D models using a structurally constrained protein evolution model.

While still a PhD student, Nico was awarded an Erasmus Mundus Sandwich PhD fellowship to spend an 8-month stay at Dr. Rita Casadio's Biocomputing Group from University of Bologna (Italy) where he started to develop structurally constrained, evolutionary-simulated Hidden Markov Models of protein families. His first Postdoctoral fellowship was awarded to take part in PhasIbeAm, the common bean genome sequencing project. Most of his work was conducted at the Protein Physiology Lab from Universidad de Buenos Aires (Argentina), where he also collaborated with Dr. Ignacio Sanchez on the application of information theory-based methods to study protein interactions by linear motifs.

Nico moved to the University of Southampton in October 2012 to work as a Research Fellow in the Edwards Lab on the project 'Integrated in silico prediction of protein-protein interaction motifs'. He left the lab in late 2014.

Employment History

• 2012-2014: Research Fellow. Centre for Biological Sciences, University of Southampton (UoS), Southampton, UK.
• 2011-2012: Postdoctoral Research Fellow. Departamento de Química Biológica, Facultad de Ciencias Exactas y Naturales, Universidad de Buenos Aires (UBA), Buenos Aires, Argentina.

Summary of Academic Qualifications

• 2011: PhD, "Development of an evolutionary-based method for the validation of protein tertiary structure and its application to starch-synthase type III from Arabidopsis thaliana." Structural Bioinformatics Group, Universidad Nacional de Quilmes (UNQ), Bernal, Buenos Aires, Argentina.
• 2006: Licentiate in Biotechnology. UNQ.

Main Funding History

• 2012-2014: Research Fellowship, UoS
• 2011-2012: Postdoctoral Fellowship, National Agency of Scientific and Technological Promotion (ANPCyT)
• 2010: Sandwich PhD Fellowship, Erasmus Mundus External Cooperation Window Lot 16
• 2008-2010: PhD Fellowship Type II, National Council for Scientific and Technical Research (CONICET)
• 2006-2008: PhD Fellowship Type I, ANPCyT