Tree-Aggregated Predictive Modeling of Microbiome Data

Bien, Jacob; Yan, Xiaohan; Simpson, Léo; Müller, Christian L.

doi:10.1101/2020.09.01.277632

Cited by 11 publications

(13 citation statements)

References 85 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The c-lasso package also integrates with R via the R package reticulate. We refer to reticulate's manual for technical details about connecting python environments and R. A successful use case of c-lasso is available in the R package trac (Bien et al, 2020), enabling tree-structured aggregation of predictors when features are rare.…”

Section: Calling C-lasso In Rmentioning

confidence: 99%

c-lasso - a Python package for constrained sparse and robust regression and classification

Simpson¹,

Combettes²,

Müller³

2021

JOSS

Self Cite

View full text Add to dashboard Cite

We introduce c-lasso, a Python package that enables sparse and robust linear regression and classification with linear equality constraints. The underlying statistical forward model is assumed to be of the following form: y = Xβ + σϵ subject to Cβ = 0 Here, X ∈ R n×d is a given design matrix and the vector y ∈ R n is a continuous or binary response vector. The matrix C is a general constraint matrix. The vector β ∈ R d contains the unknown coefficients and σ an unknown scale. Prominent use cases are (sparse) log-contrast regression with compositional data X, requiring the constraint 1 T d β = 0 (Aitchion & Bacon-Shone, 1984) and the Generalized Lasso which is a special case of the described problem (see, e.g, (James et al., 2020), Example 3). The c-lasso package provides estimators for inferring unknown coefficients and scale (i.e., perspective M-estimators (Combettes & Müller, 2020a)) of the form min β∈R d ,σ∈R0 f (Xβ − y, σ) + λ ∥β∥ 1 subject to Cβ = 0 for several convex loss functions f (•, •). This includes the constrained Lasso, the constrained scaled Lasso, sparse Huber M-estimators with linear equality constraints, and constrained (Huberized) Square Hinge Support Vector Machines (SVMs) for classification.

show abstract

Section: Calling C-lasso In Rmentioning

confidence: 99%

c-lasso - a Python package for constrained sparse and robust regression and classification

Simpson¹,

Combettes²,

Müller³

2021

JOSS

Self Cite

View full text Add to dashboard Cite

show abstract

“…We use a variant of the strategy proposed by Bien et al (2021) to make the strength of the regularization penalty dependent on the corresponding node's position in the tree. We introduce the following sigmoidal scaling:…”

Section: Node-adaptive Penalizationmentioning

confidence: 99%

“…These methods restrict themselves, however, to fully binary trees. On the other hand, the trac method (Bien et al, 2021) uses tree-guided regularization (Yan and Bien, 2021) in a maximum-likelihood-type framework to predict continuous outcomes from compositional microbiome data.…”

Section: Introductionmentioning

confidence: 99%

tascCODA: Bayesian Tree-Aggregated Analysis of Compositional Amplicon and Single-Cell Data

2021

Self Cite

View full text Add to dashboard Cite

Accurate generative statistical modeling of count data is of critical relevance for the analysis of biological datasets from high-throughput sequencing technologies. Important instances include the modeling of microbiome compositions from amplicon sequencing surveys and the analysis of cell type compositions derived from single-cell RNA sequencing. Microbial and cell type abundance data share remarkably similar statistical features, including their inherent compositionality and a natural hierarchical ordering of the individual components from taxonomic or cell lineage tree information, respectively. To this end, we introduce a Bayesian model for tree-aggregated amplicon and single-cell compositional data analysis (tascCODA) that seamlessly integrates hierarchical information and experimental covariate data into the generative modeling of compositional count data. By combining latent parameters based on the tree structure with spike-and-slab Lasso penalization, tascCODA can determine covariate effects across different levels of the population hierarchy in a data-driven parsimonious way. In the context of differential abundance testing, we validate tascCODA’s excellent performance on a comprehensive set of synthetic benchmark scenarios. Our analyses on human single-cell RNA-seq data from ulcerative colitis patients and amplicon data from patients with irritable bowel syndrome, respectively, identified aggregated cell type and taxon compositional changes that were more predictive and parsimonious than those proposed by other schemes. We posit that tascCODA1 constitutes a valuable addition to the growing statistical toolbox for generative modeling and analysis of compositional changes in microbial or cell population data.

show abstract

“…We first illustrate our microbiome knockoff generator method without phylogeny tree information on a dataset that was used to investigate the association of dietary and environmental variables with the gut microbiota. As an example of incorporating the phylogenentic structure into microbiome feature selection, we examine the HIV dataset from Bien et al (2021). The outcome of interest is soluble CD14 levels in units of pg/ml, which has been identified as a mortality predictor in HIV infection.…”

Section: Microbiome Features Predictive Of Bmi and Cd14 Levelsmentioning

confidence: 99%

Bayesian Knockoff Generators for Robust Inference Under Complex Data Structure

Märtens¹,

Banerjee²,

Qi³

et al. 2021

Preprint

View full text Add to dashboard Cite

The recent proliferation of medical data, such as genetics and electronic health records (EHR), offers new opportunities to find novel predictors of health outcomes. Presented with a large set of candidate features, interest often lies in selecting the ones most likely to be predictive of an outcome for further study such that the goal is to control the false discovery rate (FDR) at a specified level. Knockoff filtering is an innovative strategy for FDR-controlled feature selection. But, existing knockoff methods make strong distributional assumptions that hinder their applicability to real world data. We propose Bayesian models for generating high quality knockoff copies that utilize available knowledge about the data structure, thus improving the resolution of prognostic features. Applications to two feature sets are considered: those with categorical and/or continuous variables possibly having a population substructure, such as in EHR; and those with microbiome features having a compositional constraint and phylogenetic relatedness. Through simulations and real data applications, these methods are shown to identify important features with good FDR control and power.

show abstract

Tree-Aggregated Predictive Modeling of Microbiome Data

Cited by 11 publications

References 85 publications

c-lasso - a Python package for constrained sparse and robust regression and classification

c-lasso - a Python package for constrained sparse and robust regression and classification

tascCODA: Bayesian Tree-Aggregated Analysis of Compositional Amplicon and Single-Cell Data

Bayesian Knockoff Generators for Robust Inference Under Complex Data Structure

Contact Info

Product

Resources

About