A multitask transfer learning framework for the prediction of virus-human protein–protein interactions

Abstract Background Viral infections are causing significant morbidity and mortality worldwide. Understanding the interaction patterns between a particular virus and human proteins plays a crucial role in unveiling the underlying mechanism of viral infection and pathogenesis. This could further help...

Descripción completa

Guardado en:

Detalles Bibliográficos
Autores principales:	Thi Ngan Dong, Graham Brogden, Gisa Gerold, Megha Khosla
Formato:	article
Lenguaje:	EN
Publicado:	BMC 2021
Materias:	Protein–protein interaction Human PPI Virus-human PPI Multitask Transfer learning Protein embedding Computer applications to medicine. Medical informatics R858-859.7 Biology (General) QH301-705.5
Acceso en línea:	https://doaj.org/article/7f0946352b384d449aa7911036baabb1
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

id	oai:doaj.org-article:7f0946352b384d449aa7911036baabb1
record_format	dspace
spelling	oai:doaj.org-article:7f0946352b384d449aa7911036baabb12021-11-28T12:11:10ZA multitask transfer learning framework for the prediction of virus-human protein–protein interactions10.1186/s12859-021-04484-y1471-2105https://doaj.org/article/7f0946352b384d449aa7911036baabb12021-11-01T00:00:00Zhttps://doi.org/10.1186/s12859-021-04484-yhttps://doaj.org/toc/1471-2105Abstract Background Viral infections are causing significant morbidity and mortality worldwide. Understanding the interaction patterns between a particular virus and human proteins plays a crucial role in unveiling the underlying mechanism of viral infection and pathogenesis. This could further help in prevention and treatment of virus-related diseases. However, the task of predicting protein–protein interactions between a new virus and human cells is extremely challenging due to scarce data on virus-human interactions and fast mutation rates of most viruses. Results We developed a multitask transfer learning approach that exploits the information of around 24 million protein sequences and the interaction patterns from the human interactome to counter the problem of small training datasets. Instead of using hand-crafted protein features, we utilize statistically rich protein representations learned by a deep language modeling approach from a massive source of protein sequences. Additionally, we employ an additional objective which aims to maximize the probability of observing human protein–protein interactions. This additional task objective acts as a regularizer and also allows to incorporate domain knowledge to inform the virus-human protein–protein interaction prediction model. Conclusions Our approach achieved competitive results on 13 benchmark datasets and the case study for the SARS-CoV-2 virus receptor. Experimental results show that our proposed model works effectively for both virus-human and bacteria-human protein–protein interaction prediction tasks. We share our code for reproducibility and future research at https://git.l3s.uni-hannover.de/dong/multitask-transfer .Thi Ngan DongGraham BrogdenGisa GeroldMegha KhoslaBMCarticleProtein–protein interactionHuman PPIVirus-human PPIMultitaskTransfer learningProtein embeddingComputer applications to medicine. Medical informaticsR858-859.7Biology (General)QH301-705.5ENBMC Bioinformatics, Vol 22, Iss 1, Pp 1-24 (2021)
institution	DOAJ
collection	DOAJ
language	EN
topic	Protein–protein interaction Human PPI Virus-human PPI Multitask Transfer learning Protein embedding Computer applications to medicine. Medical informatics R858-859.7 Biology (General) QH301-705.5
spellingShingle	Protein–protein interaction Human PPI Virus-human PPI Multitask Transfer learning Protein embedding Computer applications to medicine. Medical informatics R858-859.7 Biology (General) QH301-705.5 Thi Ngan Dong Graham Brogden Gisa Gerold Megha Khosla A multitask transfer learning framework for the prediction of virus-human protein–protein interactions
description	Abstract Background Viral infections are causing significant morbidity and mortality worldwide. Understanding the interaction patterns between a particular virus and human proteins plays a crucial role in unveiling the underlying mechanism of viral infection and pathogenesis. This could further help in prevention and treatment of virus-related diseases. However, the task of predicting protein–protein interactions between a new virus and human cells is extremely challenging due to scarce data on virus-human interactions and fast mutation rates of most viruses. Results We developed a multitask transfer learning approach that exploits the information of around 24 million protein sequences and the interaction patterns from the human interactome to counter the problem of small training datasets. Instead of using hand-crafted protein features, we utilize statistically rich protein representations learned by a deep language modeling approach from a massive source of protein sequences. Additionally, we employ an additional objective which aims to maximize the probability of observing human protein–protein interactions. This additional task objective acts as a regularizer and also allows to incorporate domain knowledge to inform the virus-human protein–protein interaction prediction model. Conclusions Our approach achieved competitive results on 13 benchmark datasets and the case study for the SARS-CoV-2 virus receptor. Experimental results show that our proposed model works effectively for both virus-human and bacteria-human protein–protein interaction prediction tasks. We share our code for reproducibility and future research at https://git.l3s.uni-hannover.de/dong/multitask-transfer .
format	article
author	Thi Ngan Dong Graham Brogden Gisa Gerold Megha Khosla
author_facet	Thi Ngan Dong Graham Brogden Gisa Gerold Megha Khosla
author_sort	Thi Ngan Dong
title	A multitask transfer learning framework for the prediction of virus-human protein–protein interactions
title_short	A multitask transfer learning framework for the prediction of virus-human protein–protein interactions
title_full	A multitask transfer learning framework for the prediction of virus-human protein–protein interactions
title_fullStr	A multitask transfer learning framework for the prediction of virus-human protein–protein interactions
title_full_unstemmed	A multitask transfer learning framework for the prediction of virus-human protein–protein interactions
title_sort	multitask transfer learning framework for the prediction of virus-human protein–protein interactions
publisher	BMC
publishDate	2021
url	https://doaj.org/article/7f0946352b384d449aa7911036baabb1
work_keys_str_mv	AT thingandong amultitasktransferlearningframeworkforthepredictionofvirushumanproteinproteininteractions AT grahambrogden amultitasktransferlearningframeworkforthepredictionofvirushumanproteinproteininteractions AT gisagerold amultitasktransferlearningframeworkforthepredictionofvirushumanproteinproteininteractions AT meghakhosla amultitasktransferlearningframeworkforthepredictionofvirushumanproteinproteininteractions AT thingandong multitasktransferlearningframeworkforthepredictionofvirushumanproteinproteininteractions AT grahambrogden multitasktransferlearningframeworkforthepredictionofvirushumanproteinproteininteractions AT gisagerold multitasktransferlearningframeworkforthepredictionofvirushumanproteinproteininteractions AT meghakhosla multitasktransferlearningframeworkforthepredictionofvirushumanproteinproteininteractions
_version_	1718408149147320320

A multitask transfer learning framework for the prediction of virus-human protein–protein interactions

Ejemplares similares