AUTHOR=Zhang Cheng , Lei Xiujuan , Liu Lian 

TITLE=Predicting Metabolite–Disease Associations Based on LightGBM Model

JOURNAL=Frontiers in Genetics

VOLUME=Volume 12 - 2021

YEAR=2021

URL=https://www.frontiersin.org/journals/genetics/articles/10.3389/fgene.2021.660275

DOI=10.3389/fgene.2021.660275

ISSN=1664-8021

ABSTRACT=A large number of biological experiments have indicated that metabolites are closely related to the occurrence and development of many complex human diseases, which attracts many relevant researchers to excavate their correlation mechanisms. We propose a computational method named Light Gradient Boosting Machine (LightGBM) to predict potential metabolite-disease associations (LGBMMDA), which extracts features from statistical measures, graph theoretical measures, and matrix factorization results, respectively, and utilizes Principal Components Analysis (PCA) to remove the noise or redundancy. In model performance evaluation, the AUCs of LGBMMDA are better than other methods mentioned in this article. Additionally, three case studies are deeply confirmed that LGBMMDA has obvious superiority in predicting metabolite-disease pairs and is a powerful bioinformatics tool.