mbDenoise: microbiome data denoising using zero-inflated probabilistic principal components analysis

Genome Biology

卷:23

Method

Yanyan Zeng¹  Chaochun Wei¹  Jing Li¹  Tao Wang²  Hongyu Zhao³ 

[1] Department of Bioinformatics and Biostatistics, Shanghai Jiao Tong University, Shanghai, China;

[2] Department of Bioinformatics and Biostatistics, Shanghai Jiao Tong University, Shanghai, China;SJTU-Yale Joint Center for Biostatistics and Data Science, Shanghai Jiao Tong University, Shanghai, China;Department of Statistics, School of Mathematical Sciences, Shanghai Jiao Tong University, Shanghai, China;Joint International Research Laboratory of Metabolic & Developmental Sciences, Shanghai Jiao Tong University, Shanghai, China;

[3] Department of Biostatistics, Yale University, New Haven, CT, USA;SJTU-Yale Joint Center for Biostatistics and Data Science, Shanghai Jiao Tong University, Shanghai, China;

关键词: Biological zeros;    Differential abundance;    Diversity;    Negative binomial;    Normalization;   

DOI  :  10.1186/s13059-022-02657-3

 received in 2021-10-08, accepted in 2022-03-21,  发布年份 2022

来源: Springer

【 摘 要 】

The analysis of microbiome data has several technical challenges. In particular, count matrices contain a large proportion of zeros, some of which are biological, whereas others are technical. Furthermore, the measurements suffer from unequal sequencing depth, overdispersion, and data redundancy. These nuisance factors introduce substantial noise. We propose an accurate and robust method, mbDenoise, for denoising microbiome data. Assuming a zero-inflated probabilistic PCA (ZIPPCA) model, mbDenoise uses variational approximation to learn the latent structure and recovers the true abundance levels using the posterior, borrowing information across samples and taxa. mbDenoise outperforms state-of-the-art methods to extract the signal for downstream analyses.

【 授权许可】

CC BY   
© The Author(s) 2022. corrected publication 2023

【 图 表 】

13690_2023_1090_Article_IEq3.gif

Fig. 4

Fig. 1

40507_2023_167_Article_IEq517.gif

40507_2023_167_Article_IEq516.gif

Fig. 3

Fig. 2

Fig. 8

Fig. 2

Fig. 2

13638_2023_2232_Article_IEq92.gif

Fig. 2

Fig. 1

13638_2023_2232_Article_IEq89.gif

Fig. 4

Fig. 2

13638_2023_2232_Article_IEq84.gif

13638_2023_2232_Article_IEq83.gif

Fig. 7

13638_2023_2232_Article_IEq80.gif

13638_2023_2232_Article_IEq79.gif

13638_2023_2232_Article_IEq78.gif

Fig. 1

13638_2023_2232_Article_IEq75.gif

13638_2023_2232_Article_IEq73.gif

13638_2023_2232_Article_IEq72.gif

13638_2023_2232_Article_IEq71.gif

Fig. 3

Fig. 3

40507_2023_167_Article_IEq511.gif

40507_2023_167_Article_IEq507.gif

Fig. 4

40507_2023_167_Article_IEq482.gif

40507_2023_167_Article_IEq474.gif

Fig. 2

40507_2023_167_Article_IEq464.gif

40507_2023_167_Article_IEq460.gif

40507_2023_167_Article_IEq452.gif

40507_2023_167_Article_IEq447.gif