article Open AccessTop 1% cited

SignalP 6.0 predicts all five types of signal peptides using protein language models

Nature Biotechnology · 2022 · Vol. 40(7) · pp. 1023–1025

Felix Teufel✉(ETH Zurich)José Juan Almagro Armenteros(University of Copenhagen)Alexander Rosenberg Johansen(Stanford University)Magnús Halldór Gíslason(Copenhagen University Hospital)Silas Irby Pihl(Technical University of Denmark)Konstantinos D. Tsirigos(European Bioinformatics Institute)Ole Winther(University of Copenhagen)Søren Brunak(University of Copenhagen)Gunnar von Heijne(Stockholm University)Henrik Nielsen(Technical University of Denmark)

View via DOI Full-text PDF

Abstract

Signal peptides (SPs) are short amino acid sequences that control protein secretion and translocation in all living organisms. SPs can be predicted from sequence data, but existing algorithms are unable to detect all known types of SPs. We introduce SignalP 6.0, a machine learning model that detects all five SP types and is applicable to metagenomic data.

Machine Learning in Bioinformatics vaccines and immunoinformatics approaches Bioinformatics and Genomic NetworksSignal peptideSecretionSecretory proteinPeptide sequenceProtein sequencingSIGNAL (programming language)Protein Sorting Signals