CDSeq: A novel complete deconvolution method for dissecting heterogeneous samples using gene expression data

• DOI: 10.1371/journal.pcbi.1007510
• Publication Date: 2019-12-02T14:14:43Z
• Authors: Kai Kang; Qian Meng; Igor Shats; David M. Umbach; Melissa Li; Yuanyuan Li; Xiaoling Li; Leping Li
• Keywords: Cell Sorting; Cell type; Expression (computer science)

Quantifying cell-type proportions and their corresponding gene expression profiles in tissue samples would enhance understanding of the contributions individual cell types to physiological states tissue. Current approaches that address heterogeneity have drawbacks. Experimental techniques, such as fluorescence-activated sorting, single RNA sequencing are expensive. Computational use data from heterogeneous promising, but most current methods estimate either or cell-type-specific by requiring other input. Although partial deconvolution been successfully applied tumor samples, additional input required may be unavailable. We introduce a novel complete method, CDSeq, uses only RNA-Seq bulk simultaneously both profiles. Using several synthetic real experimental datasets with known composition profiles, we compared CDSeq’s performance seven established methods. Complete using CDSeq represents substantial technical advance over will useful for studying mixtures samples. is available at GitHub repository (MATLAB Octave code): https://github.com/kkang7/CDSeq.