ADAPTS: Automated Deconvolution Augmentation of Profiles for Tissue Specific cells

Abstract

Immune cell infiltration of tumors can be an important component for determining patient outcomes, e.g. by inferring immune cell presence by deconvolving gene expression data drawn from a heterogenous mix of cell types. ADAPTS aids deconvolution by adding custom cell types to existing cell-type signature matrices or building new matrices de novo. This R package builds a custom signature matrix from purified cell type gene expression data by automatically determining genes that uniquely identify each cell type. The package includes functions that call deconvolution algorithms which use the custom signature matrix to estimate the proportion of cell types present in heterogenous samples. Availability:The R packages ADAPTS, ADAPTSdata, ADAPTSdata2 are at GitHub (https://github.com/sdanzige/ADAPTS , etc.) and eventually will be available on CRAN

Bone marrow microenvironments that contribute to patient outcomes in newly diagnosed multiple myeloma: A cohort study of patients in the Total Therapy clinical trials

“…This spillover, combined with biological considerations, led to post hoc combination of several deconvolution estimates, resulting in the 18 cell types presented in this analysis. More details about signature matrix construction, including a usable version of MGSM27, are available in the Automated Deconvolution Augmentation of Profiles for Tissue Specific cells [ 19 ] package available on the Comprehensive R Archive Network ( https://cran.r-project.org/web/packages/ADAPTS/index.html ).…”

Gingival transcriptomics of follicular T cell footprints in progressing periodontitis

“…It was also predicted that certain bacterial components of the complex oral microbiome would have the capacity to aid in triggering or dampening the various molecules required for functional Tfh cells in the gingival tissues. Examination of the characteristics of gene expression representing features of Tfh cells in the gingival tissues was implemented as reported for cellular transcriptomic studies in other systems [25–30]. However, a limitation of the use of this approach to examine tissue level presentation of Tfh genes is that a number of these gene transcripts would also be produced by other cell types that reside in the gingival tissues.…”

A generalizable data-driven multicellular model of pancreatic ductal adenocarcinoma

“…Our tissue model is characterized by a set of parameters listed in Supplementary Table S1 ; some of these parameters are estimated from data available in TCGA. Specifically, cell fractions were estimated from gene expression data using “cell deconvolution” [ 54 ]. The mutation states of patient samples were summarized from a TCGA Pan-Cancer dataset [ 55 ], and deconvolved gene expression of cancer cells was generated using the DeMix algorithm [ 56 ]; see Estimation of cell fractions section for details concerning cell fraction estimation.…”

“…Cellular deconvolution [ 60 ] was used to estimate cellular fractions from bulk RNA sequencing (RNA-seq) data. In this work, we used the ADAPTS R package [ 54 ] and in particular, the SVMDECON method, which makes estimations based on support vector regression. This method solves the linear model \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{upgreek} \usepackage{mathrsfs} \setlength{\oddsidemargin}{-69pt} \begin{document} }{}$Y\ = \ AX$\end{document} , where Y is the gene expression of a given sample and A is a matrix of gene expression signatures for each cell (in columns).…”

“…However, the pancreas is composed of cell types not typically found in deconvolution resources. To create a signature matrix that includes pancreatic cells, similar to the methods found in the ADAPTS package [ 54 ], we used a pancreatic single-cell RNA sequencing (scRNA-seq) dataset in conjunction with expression signatures for 22 immune cell types (LM22) [ 61 ]. The cells found in the scRNA-seq data were previously labeled, providing a set of cells for each type.…”