Related Experiment Videos
GibbsCluster: unsupervised clustering and alignment of peptide sequences
Massimo Andreatta1, Bruno Alvarez1, Morten Nielsen1,2
1Instituto de Investigaciones Biotecnológicas, Universidad Nacional de San Martín, 1650 San Martín, Argentina.
Nucleic Acids Research
|April 14, 2017
Summary
GibbsCluster 2.0 is an improved tool for discovering sequence motifs in peptide data. It simultaneously clusters and aligns peptide sequences, identifying patterns and handling variations in motif length for better biological signaling analysis.
Area of Science:
- Bioinformatics
- Computational Biology
- Molecular Biology
Background:
- Receptor-ligand interactions are fundamental to biological signaling.
- These interactions are regulated by conserved sequence motifs within peptide ligands.
- Experimental data often contain multiple, overlapping sequence motifs.
Purpose of the Study:
- To introduce GibbsCluster 2.0, an enhanced tool for unsupervised motif discovery.
- To improve the simultaneous clustering and alignment of peptide data.
- To account for variations in motif length, including insertions and deletions.
Main Methods:
- Utilizes a clustering and alignment algorithm for peptide sequence analysis.
- Incorporates parameters for customizable cluster analysis, including penalties for small clusters and outliers.
- Applies the tool to deconvolute multiple specificities in mass spectrometry-generated peptidome data.
Main Results:
- Identifies the optimal number of clusters within peptide datasets.
- Provides sequence alignments and characteristic motifs for each identified cluster.
- Successfully deconvoluted multiple specificities in large-scale peptidome data.
Conclusions:
- GibbsCluster 2.0 is a powerful and versatile tool for motif discovery in biological sequences.
- The enhanced version effectively handles variations in motif length, improving analysis accuracy.
- Facilitates the understanding of complex receptor-ligand interactions and biological signaling pathways.