Latent variable modeling for the microbiome

Biostatistics - Tập 20 Số 4 - Trang 599-614 - 2019
Kris Sankaran1, Susan Holmes1
1Department of Statistics, Stanford University, 390 Serra Mall, Stanford, CA, USA

Tóm tắt

SummaryThe human microbiome is a complex ecological system, and describing its structure and function under different environmental conditions is important from both basic scientific and medical perspectives. Viewed through a biostatistical lens, many microbiome analysis goals can be formulated as latent variable modeling problems. However, although probabilistic latent variable models are a cornerstone of modern unsupervised learning, they are rarely applied in the context of microbiome data analysis, in spite of the evolutionary, temporal, and count structure that could be directly incorporated through such models. We explore the application of probabilistic latent variable models to microbiome data, with a focus on Latent Dirichlet allocation, Non-negative matrix factorization, and Dynamic Unigram models. To develop guidelines for when different methods are appropriate, we perform a simulation study. We further illustrate and compare these techniques using the data of Dethlefsen and Relman (2011, Incomplete recovery and individualized responses of the human distal gut microbiota to repeated antibiotic perturbation. Proceedings of the National Academy of Sciences108, 4554–4561), a study on the effects of antibiotics on bacterial community composition. Code and data for all simulations and case studies are available publicly.

Từ khóa


Tài liệu tham khảo

Blei,, 2006, Dynamic topic models., Proceedings of the 23rd International Conference on Machine Learning, 113, 10.1145/1143844.1143859

Blei,, 2003, Latent dirichlet allocation., Journal of Machine Learning Research, 3, 993

Callahan,, 2017, Exact sequence variants should replace operational taxonomic units in marker-gene data analysis., The ISME Journal, 11, 2639, 10.1038/ismej.2017.119

Canny,, 2004, Gap: a factor model for discrete data., Proceedings of the 27th Annual International ACM SIGIR Conference on Research and Development in Information Retrieval, 122

Carpenter,, 2016, Stan: a probabilistic programming language., Journal of Statistical Software, 20, 1

Chen,, 2013, Structure-constrained sparse canonical correlation analysis with an application to microbiome data analysis., Biostatistics, 14, 244, 10.1093/biostatistics/kxs038

Chen,, 2013, Variable selection for sparse Dirichlet-multinomial regression with an application to microbiome data analysis., The Annals of Applied Statistics, 7, 418, 10.1214/12-AOAS592

Chen,, 2012, Estimating functional groups in human gut microbiome with probabilistic topic models., IEEE Transactions on Nanobioscience, 11, 203, 10.1109/TNB.2012.2212204

Dethlefsen,, 2011, Incomplete recovery and individualized responses of the human distal gut microbiota to repeated antibiotic perturbation., Proceedings of the National Academy of Sciences, 108, 4554, 10.1073/pnas.1000087107

Fukuyama,, 2017, Adaptive gPCA: a method for structured dimensionality reduction., arXiv preprint arXiv:1702.00501

Fukuyama,, 2017, Multidomain analyses of a longitudinal human microbiome intestinal cleanout perturbation experiment., PLoS Computational Biology, 13, e1005706, 10.1371/journal.pcbi.1005706

Gilbert,, 2014, The earth microbiome project: successes and aspirations., BMC Biology, 12, 69, 10.1186/s12915-014-0069-1

Hoffman,, 2013, Stochastic variational inference., Journal of Machine Learning Research, 14, 1303

Holmes,, 2012, Dirichlet multinomial mixtures: generative models for microbial metagenomics., PLoS One, 7, 10.1371/journal.pone.0030126

Jiang,, 2017, Microbiome data representation by joint nonnegative matrix factorization with Laplacian regularization., IEEE/ACM Transactions on Computational Biology and Bioinformatics, 14, 353, 10.1109/TCBB.2015.2440261

Nigam,, 2000, Text classification from labeled and unlabeled documents using em., Machine Learning, 39, 103, 10.1023/A:1007692713085

Romero,, 2014, The composition and stability of the vaginal microbiota of normal pregnant women is different from that of non-pregnant women., Microbiome, 2, 4, 10.1186/2049-2618-2-4

Schloss,, 2007, The last word: books as a statistical metaphor for microbial communities., Annual Review of Microbiology, 61, 23, 10.1146/annurev.micro.61.011507.151712

Segata,, 2011, Metagenomic biomarker discovery and explanation., Genome Biology, 12, R60, 10.1186/gb-2011-12-6-r60

Shafiei,, 2015, BioMiCo: a supervised Bayesian model for inference of microbial community structure., Microbiome, 3, 8, 10.1186/s40168-015-0073-x

Teh,, 2005, Sharing clusters among related groups: Hierarchical Dirichlet processes., Advances in Neural Information Processing Systems, 1385

Wallach,, 2009, Evaluation methods for topic models., Proceedings of the 26th Annual International Conference on Machine Learning, 1105, 10.1145/1553374.1553515

Wang,, 2005, Inadequacy of interval estimates corresponding to variational Bayesian approximations. In:, AISTATS

Wang,, 2013, Nonnegative matrix factorization: a comprehensive review., IEEE Transactions on Knowledge and Data Engineering, 25, 1336, 10.1109/TKDE.2012.51

Thông tin
Thông tin xuất bản

Biostatistics

Tập 20 Số 4

599-614

Thông tin tác giả