Razor: annotation of signal peptides from toxins

Abstract

Motivation: Signal peptides are responsible for protein transport and secretion and are ubiquitous to all forms of life. The annotation of signal peptides is important for understanding protein translocation and toxin secretion, optimising recombinant protein expression, as well as for disease diagnosis and metagenomics. Results: Here we explore the features of these signal sequences across eukaryotes. We find that different kingdoms have their characteristic distributions of signal peptide residues. Additionally, the signal peptides of secretory toxins have common features across kingdoms. We leverage these subtleties to build Razor, a simple yet powerful tool for annotating signal peptides, which additionally predicts toxin- and fungal-specific signal peptides based on the first 23 N-terminal residues. Finally, we demonstrate the usability of Razor by scanning all reviewed sequences from UniProt. Indeed, Razor is able to identify toxins using their signal peptide sequences only. Strikingly, we discover that many defensive proteins across kingdoms harbour a toxin-like signal peptide; some of these defensive proteins have emerged through convergent evolution, e.g. defensin and defensin-like protein families, and phospholipase families. Availability and implementation: Razor is available as a web application (https://tisigner.com/razor) and a command-line tool (https://github.com/Gardner-BinfLab/Razor).

Antimicrobial Peptide Arsenal Predicted from the Venom Gland Transcriptome of the Tropical Trap-Jaw Ant Odontomachus chelifer

“…We worked with 125,123 predicted proteins from the transcripts assembled from the venom gland of O. chelifer . Their signal peptides were predicted using Razor v.1 [ 48 ], leaving only peptides with predicted cleavage sites and falling in the range of 8 to 98 amino acids. This precise length threshold was defined to allow the machine learning downstream operations that sometimes require peptides in these ranges.…”

Emerging Computational Approaches for Antimicrobial Peptide Discovery

“…For example, the search can be framed against those toxins/peptides having signal peptides, responsible for their transport and secretion. Signal peptides have shown to contain common features across all life kingdoms [ 171 ]. In addition, cysteine-rich secretory proteins (CRISPs), small toxins (<100 aas) commonly found within the secretions of animal venoms, can be extracted from protein databases, to enrich reference databases for increasing proteomic toxin peptides detection [ 172 ].…”