Network-based prediction of protein interactions

István A. Kovács; Katja Luck; Kerstin Spirohn; Yang Wang; Carl Pollis; Sadie Schlabach; Wenting Bian; Dae Kyum Kim; Nishka Kishore; Tong Hao; Michael A. Calderwood; Marc Vidal; Albert László Barabási

doi:10.1038/s41467-019-09177-y

Network-based prediction of protein interactions

István A. Kovács^*, Katja Luck, Kerstin Spirohn, Yang Wang, Carl Pollis, Sadie Schlabach, Wenting Bian, Dae Kyum Kim, Nishka Kishore, Tong Hao, Michael A. Calderwood, Marc Vidal, Albert László Barabási

^*Corresponding author for this work

Physics and Astronomy

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

331 Scopus citations

Abstract

Despite exceptional experimental efforts to map out the human interactome, the continued data incompleteness limits our ability to understand the molecular roots of human disease. Computational tools offer a promising alternative, helping identify biologically significant, yet unmapped protein-protein interactions (PPIs). While link prediction methods connect proteins on the basis of biological or network-based similarity, interacting proteins are not necessarily similar and similar proteins do not necessarily interact. Here, we offer structural and evolutionary evidence that proteins interact not if they are similar to each other, but if one of them is similar to the other’s partners. This approach, that mathematically relies on network paths of length three (L3), significantly outperforms all existing link prediction methods. Given its high accuracy, we show that L3 can offer mechanistic insights into disease mechanisms and can complement future experimental efforts to complete the human interactome.

Original language	English (US)
Article number	1240
Journal	Nature communications
Volume	10
Issue number	1
DOIs	https://doi.org/10.1038/s41467-019-09177-y
State	Published - Dec 1 2019

Funding

We thank D. Hill, F.P. Roth, F. Cheng and M. Santolini for useful discussions on the manuscript and A. Grishchenko for help with visualization. This work was supported by an NHGRI Center of Excellence in Genome Science grant P50HG004233 to M.V., and A.-L.B; an NHGRI U41HG001715 awarded to M.V., and M.A.C.; and an NIGMS R01GM109199 awarded to M.A.C. A.-L.B. and I.A.K were also funded by P01HL132825. We gratefully acknowledge the support of The National Human Genome Research Institute (NHGRI) and National Institute of General Medical Sciences (NIGMS) of NIH.

ASJC Scopus subject areas

General Chemistry
General Biochemistry, Genetics and Molecular Biology
General Physics and Astronomy

Access to Document

10.1038/s41467-019-09177-y

Cite this

@article{2b1a72f802e647c29fe897ad75267ee2,

title = "Network-based prediction of protein interactions",

abstract = "Despite exceptional experimental efforts to map out the human interactome, the continued data incompleteness limits our ability to understand the molecular roots of human disease. Computational tools offer a promising alternative, helping identify biologically significant, yet unmapped protein-protein interactions (PPIs). While link prediction methods connect proteins on the basis of biological or network-based similarity, interacting proteins are not necessarily similar and similar proteins do not necessarily interact. Here, we offer structural and evolutionary evidence that proteins interact not if they are similar to each other, but if one of them is similar to the other{\textquoteright}s partners. This approach, that mathematically relies on network paths of length three (L3), significantly outperforms all existing link prediction methods. Given its high accuracy, we show that L3 can offer mechanistic insights into disease mechanisms and can complement future experimental efforts to complete the human interactome.",

author = "Kov{\'a}cs, \{Istv{\'a}n A.\} and Katja Luck and Kerstin Spirohn and Yang Wang and Carl Pollis and Sadie Schlabach and Wenting Bian and Kim, \{Dae Kyum\} and Nishka Kishore and Tong Hao and Calderwood, \{Michael A.\} and Marc Vidal and Barab{\'a}si, \{Albert L{\'a}szl{\'o}\}",

note = "Publisher Copyright: {\textcopyright} 2019, The Author(s).",

year = "2019",

month = dec,

day = "1",

doi = "10.1038/s41467-019-09177-y",

language = "English (US)",

volume = "10",

journal = "Nature communications",

issn = "2041-1723",

publisher = "Nature Research",

number = "1",

}

TY - JOUR

T1 - Network-based prediction of protein interactions

AU - Kovács, István A.

AU - Luck, Katja

AU - Spirohn, Kerstin

AU - Wang, Yang

AU - Pollis, Carl

AU - Schlabach, Sadie

AU - Bian, Wenting

AU - Kim, Dae Kyum

AU - Kishore, Nishka

AU - Hao, Tong

AU - Calderwood, Michael A.

AU - Vidal, Marc

AU - Barabási, Albert László

PY - 2019/12/1

Y1 - 2019/12/1

N2 - Despite exceptional experimental efforts to map out the human interactome, the continued data incompleteness limits our ability to understand the molecular roots of human disease. Computational tools offer a promising alternative, helping identify biologically significant, yet unmapped protein-protein interactions (PPIs). While link prediction methods connect proteins on the basis of biological or network-based similarity, interacting proteins are not necessarily similar and similar proteins do not necessarily interact. Here, we offer structural and evolutionary evidence that proteins interact not if they are similar to each other, but if one of them is similar to the other’s partners. This approach, that mathematically relies on network paths of length three (L3), significantly outperforms all existing link prediction methods. Given its high accuracy, we show that L3 can offer mechanistic insights into disease mechanisms and can complement future experimental efforts to complete the human interactome.

AB - Despite exceptional experimental efforts to map out the human interactome, the continued data incompleteness limits our ability to understand the molecular roots of human disease. Computational tools offer a promising alternative, helping identify biologically significant, yet unmapped protein-protein interactions (PPIs). While link prediction methods connect proteins on the basis of biological or network-based similarity, interacting proteins are not necessarily similar and similar proteins do not necessarily interact. Here, we offer structural and evolutionary evidence that proteins interact not if they are similar to each other, but if one of them is similar to the other’s partners. This approach, that mathematically relies on network paths of length three (L3), significantly outperforms all existing link prediction methods. Given its high accuracy, we show that L3 can offer mechanistic insights into disease mechanisms and can complement future experimental efforts to complete the human interactome.

UR - http://www.scopus.com/inward/record.url?scp=85063046838&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85063046838&partnerID=8YFLogxK

U2 - 10.1038/s41467-019-09177-y

DO - 10.1038/s41467-019-09177-y

M3 - Article

C2 - 30886144

AN - SCOPUS:85063046838

SN - 2041-1723

VL - 10

JO - Nature communications

JF - Nature communications

IS - 1

M1 - 1240

ER -

Network-based prediction of protein interactions

Abstract

Funding

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this