Kejun He
Renmin University of China
28 Papers
21 Citations
Kejun He is an academic researcher from Renmin University of China. The author has contributed to research in topics: Computer science & Biology. The author has an hindex of 4, co-authored 12 publications. Previous affiliations of Kejun He include Texas A&M University.
Chat about Author
Papers
•Posted Content
LinDA: Linear Models for Differential Abundance Analysis of Microbiome Compositional Data
TL;DR: LinDA as discussed by the authors uses linear regression models on the centered log-ratio trans-formed data and corrects the bias due to compositional effects of false positive control for differentially abundance analysis of microbiome data.
181
LinDA: linear models for differential abundance analysis of microbiome compositional data
TL;DR: LinDA as mentioned in this paper uses linear regression models on the centered log-ratio transformed data, and correcting the bias due to compositional effects, which can be extended to mixed-effect models for correlated microbiome data.
Bayesian biclustering for microbial metagenomic sequencing data via multinomial matrix factorization.
TL;DR: In this article, an identifiable Bayesian multinomial matrix factorization model was proposed to infer overlapping clusters on both microbes and hosts, where the observed over-dispersed zero-inflated count matrix was represented as Dirichlet-multinomial mixtures on which latent cluster structures are built hierarchically.
21
•Posted Content
Bayesian biclustering for microbial metagenomic sequencing data via multinomial matrix factorization
TL;DR: An identifiable Bayesian multinomial matrix factorization model is proposed to infer overlapping clusters on both microbes and hosts and can help generate potential hypotheses for future investigation of the heterogeneity of the human gut microbiome.
Tensor Linear Regression: Degeneracy and Solution
TL;DR: This article provides useful results of CP degeneracy in tensor regression problems and provides a general penalized strategy as a solution to overcomeCP degeneracy.