LinDA: linear models for differential abundance analysis of microbiome compositional data

• Published in: Genome Biology Vol. 23; no. 1; p. 95
• Main Authors: Zhou, Huijuan, He, Kejun, Chen, Jun, Zhang, Xianyang
• Format: Journal Article
• Language: English
• Published: London BioMed Central 14.04.2022; BMC
• Subjects: Animal Genetics and Genomics; Animals; Bias; Bioinformatics; Biomedical and Life Sciences; Coleoptera; Compositional effect; Differential abundance analysis; Evolutionary Biology; False discovery rate; genome; Human Genetics; Life Sciences; Linear Models; Method; Methods; Microbial Genetics and Genomics; microbiome; Microbiomes; Microbiota; Microorganisms; Multiple testing; Normal distribution; Numerical analysis; Performance evaluation; Plant Genetics and Genomics; Regression analysis; Research Design; Statistical analysis; Taxonomy; False discovery rate; Differential abundance analysis; Multiple testing; Compositional effect
• ISSN: 1474-760X; 1474-7596; 1474-760X
• DOI: 10.1186/s13059-022-02655-5

Differential abundance analysis is at the core of statistical analysis of microbiome data. The compositional nature of microbiome sequencing data makes false positive control challenging. Here, we show that the compositional effects can be addressed by a simple, yet highly flexible and scalable, approach. The proposed method, LinDA, only requires fitting linear regression models on the centered log-ratio transformed data, and correcting the bias due to compositional effects. We show that LinDA enjoys asymptotic FDR control and can be extended to mixed-effect models for correlated microbiome data. Using simulations and real examples, we demonstrate the effectiveness of LinDA.