Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2018 May 15;34(10):1690-1696.
doi: 10.1093/bioinformatics/btx818.

DeepSig: deep learning improves signal peptide detection in proteins

Castrense Savojardo  1 , Pier Luigi Martelli  1 , Piero Fariselli  2 , Rita Casadio  1
Affiliations

DeepSig: deep learning improves signal peptide detection in proteins

Castrense Savojardo et al. Bioinformatics. .

Abstract

Motivation: The identification of signal peptides in protein sequences is an important step toward protein localization and function characterization.

Results: Here, we present DeepSig, an improved approach for signal peptide detection and cleavage-site prediction based on deep learning methods. Comparative benchmarks performed on an updated independent dataset of proteins show that DeepSig is the current best performing method, scoring better than other available state-of-the-art approaches on both signal peptide detection and precise cleavage-site identification.

Availability and implementation: DeepSig is available as both standalone program and web server at https://deepsig.biocomp.unibo.it. All datasets used in this study can be obtained from the same website.

Contact: pierluigi.martelli@unibo.it.

Supplementary information: Supplementary data are available at Bioinformatics online.

PubMed Disclaimer

Figures

Fig. 1.
Fig. 1.
The architecture of the DCNN processing an input protein sequence to detect signal peptides. Feature extraction involves the application of three convolution-pooling (conv-pool) stages. The final classification is performed by a standard fully-connected neural network
Fig. 2.
Fig. 2.
The signal-peptide GRHCRF model capturing the modular structure of the signal peptide. States labeled with N, H, and C represents the positively charged N-region, the hydrophobic H-region and the cleavage C-region, respectively (see Section 2.4 for further details)
See this image and copyright information in PMC

References

    1. Abadi M. et al. (2015) Tensorflow: Large-Scale Machine Learning on Heterogeneous Systems. Software available online: https://www.tensorflow.org.
    1. Alipanahi B. et al. (2015) Predicting the sequence specificities of DNA- and RN-binding proteins by deep learning. Nat. Biotechnol., 33, 831–838. - PubMed
    1. Bach S. et al. (2015) On pixel-wise explanations for non-linear classifier decision by layer-wise relevance propagation. PLoS One, 10, e0130140.. - PMC - PubMed
    1. Bagos P.G. et al. (2010) Combined prediction of Tat and Sec signal peptides with hidden Markov models. Bioinformatics, 26, 2811–2817. - PubMed
    1. Berks B.C. (2015) The twin-arginine protein translocation pathway. Annu. Rev. Biochem., 84, 843–864. - PubMed

Substances