Challenges and opportunities to computationally deconvolve heterogeneous tissue with varying cell sizes using single-cell RNA-sequencing datasets

Abstract

Deconvolution of cell mixtures in “bulk” transcriptomic samples from homogenate human tissue is important for understanding disease pathologies. However, several experimental and computational challenges impede transcriptomics-based deconvolution approaches using single-cell/nucleus RNA-seq reference atlases. Cells from the brain and blood have substantially different sizes, total mRNA, and transcriptional activities, and existing approaches may quantify total mRNA instead of cell type proportions. Further, standards are lacking for the use of cell reference atlases and integrative analyses of single-cell and spatial transcriptomics data. We discuss how to approach these key challenges with orthogonal “gold standard” datasets for evaluating deconvolution methods.

DISSECT: deep semi-supervised consistency regularization for accurate cell type fraction and gene expression estimation

“…However, is unobserved, and the deconvolution algorithm is learned using simulations. This learning process involving simulations is highly dependent on the reference being the single-cell dataset used to generate simulations, and is subjected to an inherent strong domain shift [ 14 ]. To address this, we hypothesize that a consistency-based regularization penalizing the non-linearity of mixtures of real and simulated samples would result in a mapping that is closer to true mapping f .…”

Advances and applications of biomimetic biomaterials for endogenous skin regeneration

“…For example, errors due to the cell extraction process, abundance, size and total mRNA of the cells were difficult to avoid and could not be verified by setting up orthogonal experiments. Besides, lacking of standardization of cell type annotation and marker selection strategies also make it difficult to achieve efficient data processing [ 299 ]. But as emerging technologies in recent years, it is believed that single-cell sequencing and spatial transcriptomics technologies will become representative of histologic technologies in the future to facilitate skin endogenous regeneration.…”

CATD: a reproducible pipeline for selecting cell-type deconvolution methods across tissues

“…Therefore, a centralized and efficient platform is essential to streamline the evaluation process, facilitating a quicker and more unified assessment of deconvolution methods. Furthermore, while previous studies have primarily focused on deconvolution in peripheral blood mononuclear cell (PBMC) samples or in a limited number of tissues each time, there is a growing demand for research that covers more tissues and datasets as more and more cell type-specific information becomes available ( Maden et al 2023 ).…”