Predicting host tropism of influenza A virus proteins using random forest

Abstract

BackgroundMajority of influenza A viruses reside and circulate among animal populations, seldom infecting humans due to host range restriction. Yet when some avian strains do acquire the ability to overcome species barrier, they might become adapted to humans, replicating efficiently and causing diseases, leading to potential pandemic. With the huge influenza A virus reservoir in wild birds, it is a cause for concern when a new influenza strain emerges with the ability to cross host species barrier, as shown in light of the recent H7N9 outbreak in China. Several influenza proteins have been shown to be major determinants in host tropism. Further understanding and determining host tropism would be important in identifying zoonotic influenza virus strains capable of crossing species barrier and infecting humans.ResultsIn this study, computational models for 11 influenza proteins have been constructed using the machine learning algorithm random forest for prediction of host tropism. The prediction models were trained on influenza protein sequences isolated from both avian and human samples, which were transformed into amino acid physicochemical properties feature vectors. The results were highly accurate prediction models (ACC>96.57; AUC>0.980; MCC>0.916) capable of determining host tropism of individual influenza proteins. In addition, features from all 11 proteins were used to construct a combined model to predict host tropism of influenza virus strains. This would help assess a novel influenza strain's host range capability.ConclusionsFrom the prediction models constructed, all achieved high prediction performance, indicating clear distinctions in both avian and human proteins. When used together as a host tropism prediction system, zoonotic strains could potentially be identified based on different protein prediction results. Understanding and predicting host tropism of influenza proteins lay an important foundation for future work in constructing computation models capable of directly predicting interspecies transmission of influenza viruses. The models are available for prediction at http://fluleap.bic.nus.edu.sg.

Natural Reassortants of Potentially Zoonotic Avian Influenza Viruses H5N1 and H9N2 from Egypt Display Distinct Pathogenic Phenotypes in Experimentally Infected Chickens and Ferrets

“…The sequences of 11 viral proteins translated from the genomic sequences of the two strains (HA, M1, M2, NA, NP, NS1, NEP, PA, PB1, PB1-F2, and PB2) were first subjected to individual host tropism predictions. The protein sequences, represented by amino acid compositions and physicochemical properties, were individually predicted to have either avian or human protein tropism using the random forest machine learning classifier (49). The host tropism protein prediction results for the 11 proteins of each strain were then combined as a host tropism protein signature, which illustrates the underlying avian or human tropism of the strains (24).…”

HopPER: an adaptive model for probability estimation of influenza reassortment through host prediction

“…We first carry out the experiments on the host tropism prediction for selected proteins. The effectiveness of host tropism prediction on influenza HA proteins and zoonotic strains prediction has been demonstrated by Eng et al [19,27]. Our previous work supplemented this work on the host prediction of human-adapted subtypes using random forest that achieved better results over other classifiers [28].…”

The Central Role of Non-Structural Protein 1 (NS1) in Influenza Biology and Infection

“…For example, random forests (RF) was used to build 11 computational models from sequences of proteins isolated from influenza, producing highly accurate prediction models capable of determining the host tropism of individual influenza proteins. In this study, NS1 was predicted to have a large number of interactions with different hosts (high host tropism) [161].…”