Skip to main content
Journal cover image

Multisample estimation of bacterial composition matrices in metagenomics data

Publication ,  Journal Article
Cao, Y; Zhang, A; Li, H
Published in: Biometrika
March 1, 2020

Metagenomics sequencing is routinely applied to quantify bacterial abundances in microbiome studies, where bacterial composition is estimated based on the sequencing read counts. Due to limited sequencing depth and DNA dropouts, many rare bacterial taxa might not be captured in the final sequencing reads, which results in many zero counts. Naive composition estimation using count normalization leads to many zero proportions, which tend to result in inaccurate estimates of bacterial abundance and diversity. This paper takes a multisample approach to estimation of bacterial abundances in order to borrow information across samples and across species. Empirical results from real datasets suggest that the composition matrix over multiple samples is approximately low rank, which motivates a regularized maximum likelihood estimation with a nuclear norm penalty. An efficient optimization algorithm using the generalized accelerated proximal gradient and Euclidean projection onto simplex space is developed. Theoretical upper bounds and the minimax lower bounds of the estimation errors, measured by the Kullback–Leibler divergence and the Frobenius norm, are established. Simulation studies demonstrate that the proposed estimator outperforms the naive estimators. The method is applied to an analysis of a human gut microbiome dataset.

Duke Scholars

Published In

Biometrika

DOI

EISSN

1464-3510

ISSN

0006-3444

Publication Date

March 1, 2020

Volume

107

Issue

1

Start / End Page

75 / 92

Publisher

Oxford University Press (OUP)

Related Subject Headings

  • Statistics & Probability
  • 1403 Econometrics
  • 0104 Statistics
  • 0103 Numerical and Computational Mathematics
 

Citation

APA
Chicago
ICMJE
MLA
NLM
Cao, Y., Zhang, A., & Li, H. (2020). Multisample estimation of bacterial composition matrices in metagenomics data. Biometrika, 107(1), 75–92. https://doi.org/10.1093/biomet/asz062
Cao, Yuanpei, Anru Zhang, and Hongzhe Li. “Multisample estimation of bacterial composition matrices in metagenomics data.” Biometrika 107, no. 1 (March 1, 2020): 75–92. https://doi.org/10.1093/biomet/asz062.
Cao Y, Zhang A, Li H. Multisample estimation of bacterial composition matrices in metagenomics data. Biometrika. 2020 Mar 1;107(1):75–92.
Cao, Yuanpei, et al. “Multisample estimation of bacterial composition matrices in metagenomics data.” Biometrika, vol. 107, no. 1, Oxford University Press (OUP), Mar. 2020, pp. 75–92. Crossref, doi:10.1093/biomet/asz062.
Cao Y, Zhang A, Li H. Multisample estimation of bacterial composition matrices in metagenomics data. Biometrika. Oxford University Press (OUP); 2020 Mar 1;107(1):75–92.
Journal cover image

Published In

Biometrika

DOI

EISSN

1464-3510

ISSN

0006-3444

Publication Date

March 1, 2020

Volume

107

Issue

1

Start / End Page

75 / 92

Publisher

Oxford University Press (OUP)

Related Subject Headings

  • Statistics & Probability
  • 1403 Econometrics
  • 0104 Statistics
  • 0103 Numerical and Computational Mathematics