Abstract
Metabolites play a significant role in various complex human disease. The exploration of the relationship between metabolites and diseases can help us to better understand the underlying pathogenesis. Several network-based methods have been used to predict the association between metabolite and disease. However, some methods ignored hierarchical differences in disease network and failed to work in the absence of known metabolite-disease associations. This paper presents a bi-random walks based method for disease-related metabolites prediction, called MDBIRW. First of all, we reconstruct the disease similarity network and metabolite functional similarity network by integrating Gaussian Interaction Profile (GIP) kernel similarity of diseases and GIP kernel similarity of metabolites, respectively. Then, the bi-random walks algorithm is executed on the reconstructed disease similarity network and metabolite functional similarity network to predict potential disease-metabolite associations. At last, MDBIRW achieves reliable performance in leave-one-out cross validation (AUC of 0.910) and 5-fold cross validation (AUC of 0.924). The experimental results show that our method outperforms other existing methods for predicting disease-related metabolites.
Highlights
Metabolites play an important role in the maintenance, growth and reproduction of organisms, and are greatly helpful to illustrate the underlying molecular disease-causing mechanisms [1]
Czech et al used the method of gas and liquid chromatography-tandem mass spectrometry (GC-MS and LC-MS/MS) to analyze CSF samples in Alzheimer’s
We downloaded the metabolites data and diseases data from Human Metabolome Database (HMDB) [17] and Human Disease Ontology (DO) [18], respectively (S1 File). 2262 metabolites, 216 diseases and 4537 metabolite-disease associations can be obtained after removing redundant associations
Summary
Metabolites play an important role in the maintenance, growth and reproduction of organisms, and are greatly helpful to illustrate the underlying molecular disease-causing mechanisms [1]. There is abundant evidence that diseases are always accompanied with changes in metabolite [2]. It is significant to identify abnormal metabolites for diagnosis and treatment of diseases [3]. As the development of molecular technology, many researchers have revealed the association between disease and other molecular products like gene, microRNA, circRNA, protein, etc [4,5,6]. Luo et al used BIRW to predict the potential association between drug and disease [7]. Yan et al developed the method DNRLMF-MDA by integrating disease similarity and miRNA similarity to predict disease-related miRNA based on dynamic neighbourhood regularized logistic matrix factorization [8]. Czech et al used the method of gas and liquid chromatography-tandem mass spectrometry (GC-MS and LC-MS/MS) to analyze CSF samples in Alzheimer’s
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
