GibbsCluster: unsupervised clustering and alignment of peptide sequences

Abstract

Receptor interactions with short linear peptide fragments (ligands) are at the base of many biological signaling processes. Conserved and information-rich amino acid patterns, commonly called sequence motifs, shape and regulate these interactions. Because of the properties of a receptor-ligand system or of the assay used to interrogate it, experimental data often contain multiple sequence motifs. GibbsCluster is a powerful tool for unsupervised motif discovery because it can simultaneously cluster and align peptide data. The GibbsCluster 2.0 presented here is an improved version incorporating insertion and deletions accounting for variations in motif length in the peptide input. In basic terms, the program takes as input a set of peptide sequences and clusters them into meaningful groups. It returns the optimal number of clusters it identified, together with the sequence alignment and sequence motif characterizing each cluster. Several parameters are available to customize cluster analysis, including adjustable penalties for small clusters and overlapping groups and a trash cluster to remove outliers. As an example application, we used the server to deconvolute multiple specificities in large-scale peptidome data generated by mass spectrometry. The server is available at http://www.cbs.dtu.dk/services/GibbsCluster-2.0.

MS‐Rescue: A Computational Pipeline to Increase the Quality and Yield of Immunopeptidomics Experiments

“…The GibbsCluster analysis detected the correct number of specificities for the TC‐1 set and A18‐Tpm, while it underestimated the number of alleles in the Jurkat data set (Figure S1, Supporting Information). As has been previously reported, unsupervised clustering tends to underestimate the number of specificities when multiple MHCs have redundant motifs and/or low expression levels . In the case of the Jurkat cell line, the expressed HLA‐B alleles B*07:02 and B*35:03 belong to the B07 supertype and are known to have nearly identical peptide‐binding preferences; unsupervised clustering, therefore, fails to separate them into two separate specificities.…”

“…As has been previously reported, unsupervised clustering tends to underestimate the number of specificities when multiple MHCs have redundant motifs and/or low expression levels. [16,22] In the case of the Jurkat cell line, the expressed HLA-B alleles B*07:02 and B*35:03 belong to the B07 supertype and are known to have nearly identical peptide-binding preferences [23,24] ; unsupervised clustering, therefore, fails to separate them into two separate specificities. Finally, the fraction of peptides captured by the "trash cluster" varied between 2.6% (A18-Tpm) and 7.9% (TC-1), suggesting a level of variability in the quality of the different data sets.…”

NetMHCpan-4.0: Improved Peptide–MHC Class I Interaction Predictions Integrating Eluted Ligand and Peptide Binding Affinity Data

“…The only critical limitation for such data integrations is the criteria that each data point must be associated with a specific MHC element. This information is not always readily available, but can in most cases be inferred by unsupervised clustering of the available data (using GibbsCluster (29), position weight matrix mixture models (16), or similar approaches), and association of each cluster to an MHC molecule of the given host.…”

Development of a novel clustering tool for linear peptide sequences

“…For example, Hammock 23 was developed to identify consensus motifs in large data sets, and PepServe 24 to cluster peptides based on their physiochemical properties. Another application, UCLUST, 25 performs rapid clustering and implements several previously identified clustering algorithms, and Andreatta et al 26,27 developed a tool based on Gibbs sampling. Overall, most of these tools cluster based on the physiochemical properties of peptide sequences or by defining and focusing on shared motifs.…”