SpotOn: High Accuracy Identification of Protein-Protein Interface Hot-Spots

Irina de Sousa Moreira; Panos Koukos; Rita Melo; Jose G Almeida; Antonio J. Preto; Joerg Schaarschmidt; Mikael Trellet; Zeynep H Gümüş; Joaquim Costa; Alexandre M. J. J. Bonvin

doi:10.1038/s41598-017-08321-2

SpotOn: High Accuracy Identification of Protein-Protein Interface Hot-Spots

Irina de Sousa Moreira, Panos Koukos, Rita Melo, Jose G Almeida, Antonio J. Preto, Joerg Schaarschmidt, Mikael Trellet, Zeynep H Gümüş, Joaquim Costa, Alexandre M. J. J. Bonvin

University of Coimbra

Research output: Contribution to journal › Article › Academic › peer-review

Abstract

We present SpotOn, a web server to identify and classify interfacial residues as Hot-Spots (HS) and Null-Spots (NS). SpotON implements a robust algorithm with a demonstrated accuracy of 0.95 and sensitivity of 0.98 on an independent test set. The predictor was developed using an ensemble machine learning approach with up-sampling of the minor class. It was trained on 53 complexes using various features, based on both protein 3D structure and sequence. The SpotOn web interface is freely available at: http://milou.science.uu.nl/services/SPOTON/.

Original language	English
Article number	8007
Journal	Scientific Reports
Volume	7
Issue number	1
DOIs	https://doi.org/10.1038/s41598-017-08321-2
Publication status	Published - 1 Dec 2017

Access to Document

10.1038/s41598-017-08321-2

s41598-017-08321-2Final published version, 2.56 MB

http://www.nature.com/articles/s41598-017-08321-2

Cite this

@article{cab8eafb01934127a843298845fb63a4,

title = "SpotOn: High Accuracy Identification of Protein-Protein Interface Hot-Spots",

abstract = "We present SpotOn, a web server to identify and classify interfacial residues as Hot-Spots (HS) and Null-Spots (NS). SpotON implements a robust algorithm with a demonstrated accuracy of 0.95 and sensitivity of 0.98 on an independent test set. The predictor was developed using an ensemble machine learning approach with up-sampling of the minor class. It was trained on 53 complexes using various features, based on both protein 3D structure and sequence. The SpotOn web interface is freely available at: http://milou.science.uu.nl/services/SPOTON/.",

author = "{de Sousa Moreira}, Irina and Panos Koukos and Rita Melo and Almeida, {Jose G} and Preto, {Antonio J.} and Joerg Schaarschmidt and Mikael Trellet and G{\"u}m{\"u}{\c s}, {Zeynep H} and Joaquim Costa and Bonvin, {Alexandre M. J. J.}",

year = "2017",

month = dec,

day = "1",

doi = "10.1038/s41598-017-08321-2",

language = "English",

volume = "7",

journal = "Scientific Reports",

issn = "2045-2322",

publisher = "NLM (Medline)",

number = "1",

}

TY - JOUR

T1 - SpotOn: High Accuracy Identification of Protein-Protein Interface Hot-Spots

AU - de Sousa Moreira, Irina

AU - Koukos, Panos

AU - Melo, Rita

AU - Almeida, Jose G

AU - Preto, Antonio J.

AU - Schaarschmidt, Joerg

AU - Trellet, Mikael

AU - Gümüş, Zeynep H

AU - Costa, Joaquim

AU - Bonvin, Alexandre M. J. J.

PY - 2017/12/1

Y1 - 2017/12/1

N2 - We present SpotOn, a web server to identify and classify interfacial residues as Hot-Spots (HS) and Null-Spots (NS). SpotON implements a robust algorithm with a demonstrated accuracy of 0.95 and sensitivity of 0.98 on an independent test set. The predictor was developed using an ensemble machine learning approach with up-sampling of the minor class. It was trained on 53 complexes using various features, based on both protein 3D structure and sequence. The SpotOn web interface is freely available at: http://milou.science.uu.nl/services/SPOTON/.

AB - We present SpotOn, a web server to identify and classify interfacial residues as Hot-Spots (HS) and Null-Spots (NS). SpotON implements a robust algorithm with a demonstrated accuracy of 0.95 and sensitivity of 0.98 on an independent test set. The predictor was developed using an ensemble machine learning approach with up-sampling of the minor class. It was trained on 53 complexes using various features, based on both protein 3D structure and sequence. The SpotOn web interface is freely available at: http://milou.science.uu.nl/services/SPOTON/.

U2 - 10.1038/s41598-017-08321-2

DO - 10.1038/s41598-017-08321-2

M3 - Article

SN - 2045-2322

VL - 7

JO - Scientific Reports

JF - Scientific Reports

IS - 1

M1 - 8007

ER -

SpotOn: High Accuracy Identification of Protein-Protein Interface Hot-Spots

Abstract

Access to Document

Fingerprint

Cite this