Model-Based Microbiome Data Ordination: A Variational Approximation Approach. Issue 4 (2nd October 2021)

• Record Type: Journal Article
• Authors: Zeng, Yanyan; Zhao, Hongyu; Wang, Tao
• Is Part Of: Journal of computational and graphical statistics. Volume 30:Issue 4(2021)
• Page Start: 1036
• Page End: 1048
• Publication Date: 2021-10-02
• Subjects: Correlated counts -- Negative binomial -- Over-dispersion -- Principal coordinate analysis -- Variational inference -- Zero-inflation; Mathematical statistics -- Data processing -- Periodicals; Mathematical statistics -- Graphic methods -- Periodicals; 519.50285
• DOI: 10.1080/10618600.2021.1882467
• Languages: English
• ISSNs: 1061-8600

Abstract

Abstract: The coevolution between human and bacteria colonizing the human body has profound implications for heath and development, with a growing body of evidence linking the altered microbiome composition with a wide array of disease states. Yet dimension reduction and visualization analysis of microbiome data are still in their infancy and many challenges exist. In this article, we introduce a general framework, zero-inflated probabilistic principal component analysis (ZIPPCA), for dimension reduction and data ordination of multivariate abundance data, and propose an efficient variational approximation method for estimation, inference, and prediction. Extensive simulations show that the proposed method outperforms algorithm-based methods and compares favorably with existing model-based methods. We further apply our method to a gut microbiome dataset for visualization analysis of community composition across age and geography. The method is implemented in R and available at https://github.com/YanyZeng/ZIPPCA .