Predicting Microbe-Disease Association by Kernelized Bayesian Matrix Factorization

The study of microbe-disease associations can be utilized as a valuable material for understanding disease pathogenesis. Developing a highly accurate algorithm model for predicting disease-related microbes will provide a basis for targeted treatment of the disease. In this paper, we propose an approach based on Kernelized Bayesian Matrix Factorization (KBMF) to predict microbe-disease association, based on the Gaussian interaction profile kernel similarity for microbes and diseases. The prediction performance of the method was evaluated by five-fold cross validation. KBMF achieved reliable results which is better than several state-of-the-art methods with around 8% improvement of AUC. Furthermore, case studies have demonstrated the reliability of the method.

[1]  Xing Chen,et al.  Prediction of microbe–disease association from the integration of neighbor and graph with collaborative recommendation model , 2017, Journal of Translational Medicine.

[2]  Luiz R Travassos,et al.  Treatment with Propionibacterium acnes modulates the late phase reaction of immediate hypersensitivity in mice. , 2003, Immunology letters.

[3]  Gregor Reid,et al.  The influence of the human microbiome and probiotics on cardiovascular health , 2014, Gut microbes.

[4]  Marcus J. Claesson,et al.  Genome-scale analyses of health-promoting bacteria: probiogenomics , 2009, Nature Reviews Microbiology.

[5]  Zhu-Hong You,et al.  PBHMDA: Path-Based Human Microbe-Disease Association Prediction , 2017, Front. Microbiol..

[6]  Dean Billheimer,et al.  Asthma-associated differences in microbial composition of induced sputum. , 2013, The Journal of allergy and clinical immunology.

[7]  S. Taylor-Robinson,et al.  The promise of metabolic phenotyping in gastroenterology and hepatology , 2015, Nature Reviews Gastroenterology &Hepatology.

[8]  Mehmet Gönen,et al.  Predicting drug-target interactions from chemical and genomic kernels using Bayesian matrix factorization , 2012, Bioinform..

[9]  Kevin Kennedy,et al.  The Home Microbiome and Childhood Asthma , 2014 .

[10]  Chunyan Miao,et al.  Neighborhood Regularized Logistic Matrix Factorization for Drug-Target Interaction Prediction , 2016, PLoS Comput. Biol..

[11]  Matthew J. Beal.  Variational algorithms for approximate Bayesian inference , 2003 .

[12]  Byoung-Ju Kim,et al.  The Effects of Lactobacillus rhamnosus on the Prevention of Asthma in a Murine Model , 2010, Allergy, asthma & immunology research.

[13]  Xing Chen,et al.  Novel human lncRNA-disease association inference based on lncRNA expression profiles , 2013, Bioinform..

[14]  Jingpu Zhang,et al.  A novel approach for predicting microbe-disease associations by bi-random walk on the heterogeneous network , 2017, PLoS ONE.

[15]  Xianjun Shen,et al.  Predicting disease-microbe association by random walking on the heterogeneous network , 2016, 2016 IEEE International Conference on Bioinformatics and Biomedicine (BIBM).