Dataset Information

A novel computational complete deconvolution method using RNA-seq data

ABSTRACT: The cell type composition of many biological tissues varies widely across samples. Such sample heterogeneity hampers efforts to probe the role of each cell type in the tissue microenvironment. Current approaches that address this issue have drawbacks. Cell sorting or single-cell based experimental techniques disrupt in situ interactions and alter physiological status of cells in tissues. Computational methods are flexible and promising; but they often estimate either sample-specific proportions of each cell type or cell-type-specific gene expression profiles, not both, by requiring the other as input. We introduce a computational Complete Deconvolution method that can estimate both sample-specific proportions of each cell type and cell-type-specific gene expression profiles simultaneously using bulk RNA-Seq data only (CDSeq). We assessed our method’s performance using several synthetic and experimental mixtures of varied but known cell-type composition and compared its performance to the performance of two state-of-the-art deconvolution methods on the same mixtures. The results showed CDSeq can estimate both sample-specific proportions of each component cell type and cell-type-specific gene expression profiles with high accuracy. CDSeq holds promise for computationally deciphering complex mixtures of cell types, each with differing expression profiles, using RNA-seq data measured in bulk tissue .

ORGANISM(S): Homo sapiens

PROVIDER: GSE123604 | GEO | 2018/12/12

REPOSITORIES: GEO

ACCESS DATA

Similar Datasets

Project description:DNA methylation assessments of peripheral blood DNA can be used to accurately estimate the relative proportions of underlying leukocyte subtypes. Such cell deconvolution analysis relies on libraries of discriminating differentially methylated regions that are developed for each specific cell type measured. The relationship between estimated cell type proportions can then be tested for their association with phenotypes, disease states, and subject outcomes, or used in multivariable models as terms for adjustment in epigenome-wide association studies (EWAS). We obtained purified neutrophils, monocytes, B-lymphocytes, natural killer (NK) cells, CD4+ T-cells, and CD8+ T-cells from healthy subjects and measured DNA methylation with the Illumina HumanMethylationEPIC array platform. In addition, we measured DNA methylation with the EPIC array in two sets of artificial DNA mixtures comprising the above cell types. We compared three separate approaches to select reference differentially methylated region libraries (DMR library), for cell type proportion inference. The IDOL algorithm identified an optimal DMR library consisting of 450 CpG sites for inferring leukocyte subtype proportions (average R2=99.2). Importantly, the majority of CpG sites (69%) in the IDOL DMR library were unique to the new EPIC methylation array, in that they were not present on the 450K array. Our new reference DMR library is available as a Bioconductor package, has the potential to reduce any unintended technical differences arising from the combination of different generations of array platforms, and may be helpful in generating larger DMR libraries that include novel cell subtypes. A longitudinal dataset of 12 measurements in a single healthy subject (age 33 at the beginning of the blood collection) was used to apply our new DMR library. The subject was followed up for approximately 400 days with samples collected every 40 days. One of the samples was excluded from the final analysis due to a potential mix-up.

Dataset Information

A novel computational complete deconvolution method using RNA-seq data

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure