Skip to main content
ARS Home » Plains Area » Clay Center, Nebraska » U.S. Meat Animal Research Center » Genetics and Animal Breeding » Research » Publications at this Location » Publication #405533

Research Project: Multi-Dimension Phenotyping to Enhance Prediction of Performance in Swine

Location: Genetics and Animal Breeding

Title: Estimation of cell type proportions from bulk RNA-Seq of porcine whole blood samples using partial reference-free deconvolution

Author
item Keel, Brittney
item Lindholm-Perry, Amanda
item Rohrer, Gary
item Oliver, William

Submitted to: Animal Gene
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 9/11/2023
Publication Date: 9/16/2023
Citation: Keel-Mercer, B.N., Lindholm-Perry, A.K., Rohrer, G.A., Oliver, W.T. 2023. Estimation of cell type proportions from bulk RNA-Seq of porcine whole blood samples using partial reference-free deconvolution. Animal Gene. 30. Article 200159. https://doi.org/10.1016/j.angen.2023.200159.
DOI: https://doi.org/10.1016/j.angen.2023.200159

Interpretive Summary: Whole blood has become increasingly utilized in RNA-sequencing (RNA-Seq) analysis because it is easily accessible and can be collected from live animals with minimal invasiveness. However, whole blood represents an extremely complex mixture of cell types, and differences in cell types can confound statistical analyses of the data. Experimental approaches for cell counting, such as cell sorting, are arduous and expensive. Statistical approaches have been developed to estimate cell type proportions directly from RNA-Seq data. In addition to being financially advantageous, statistical estimation can readily be applied to old datasets, where it may be difficult or impossible to re-analyze for cell type information. Statistical estimation procedures require a reference data set for the cell types of interest. In pigs, reference data is available for only a small number of blood cell types. ARS researchers have developed a manually curated reference set of porcine blood cell markers using publicly available swine genomics data and complementary data from human and mouse. The reference data set can be utilized to obtain estimates of cell type proportions of neutrophils, lymphocytes, monocytes, eosinophils, basophils, and red blood cells. Estimates for these cell types can then be used in downstream statistical models to adjust for potential confounding effects of cell composition. This reference set will serve as a valuable resource for estimating cell proportions from porcine whole blood when other quantitative measurements are unavailable.

Technical Abstract: Whole blood has become increasingly utilized in transcriptomic studies because it is easily accessible and can be collected from live animals with minimal invasiveness. However, whole blood represents an extremely complex mixture of cell types, and cell type proportions can confound downstream statistical analyses. Information on cell type proportions may be missing from blood transcriptome studies for a variety of reasons. Experimental approaches for cell counting, such as cell sorting, are arduous and expensive, and therefore may not feasible for studies conducted on a limited budget. Statistical deconvolution can be applied directly to transcriptomic data sets to estimate cell type proportions. In addition to being financially advantageous, computational deconvolution can readily be applied to old datasets, where it may be difficult or impossible to re-analyze for cell type information. In an effort to assist researchers in recovering cell type proportions from porcine whole blood transcriptome samples, we present a manually curated set of porcine blood cell markers that can be utilized in a partial reference-free deconvolution framework to obtain estimates of cell types measured in a standard complete blood count (CBC) panel, which includes neutrophils, lymphocytes, monocytes, eosinophils, basophils, and red blood cells.