# Optimizing single cell RNA sequencing of stem cells. A streamlined workflow for enhanced sensitivity and reproducibility in hematopoietic studies. The use of human umbilical cord blood-derived hematopoietic stem and progenitor cells

## TL;DR

This paper presents an optimized workflow for single-cell RNA sequencing of hematopoietic stem cells from umbilical cord blood, improving sensitivity and reproducibility.

## Contribution

A streamlined and validated scRNA-seq protocol for analyzing limited HSPC populations from umbilical cord blood.

## Key findings

- CD34+Lin−CD45+ and CD133+Lin−CD45+ HSPCs show very similar gene expression profiles (R = 0.99).
- Integrated pseudobulk analysis of both cell types enhances the reliability of scRNA-seq results.
- The workflow confirms the feasibility of HSPC analysis with small cell numbers.

## Abstract

Human hematopoietic stem/progenitor cells (HSPCs) are enriched in umbilical cord blood (UCB) among cell populations that express CD34 and CD133 (PROM1) antigens. These cells can be purified further and sorted by FACS as CD34+Lin−CD45+ and CD133+Lin−CD45+ cells. It has been postulated that the population of CD133+ HSPCs is enriched for more primitive stem cells. To address this issue at the molecular level, we performed single-cell RNA-sequencing (scRNA-seq) and analyzed the transcriptome of both cell types. We optimized the available protocols of scRNA-seq of HSPC and described our laboratory experiences with the limited number of cells obtained from human UCB.

Herein, we report the results of scRNA-seq analysis paying special attention to the quality parameters of single cell libraries. We also present the similarities and differences in transcriptome between these cells (CD34+Lin-CD45+ and CD133+Lin-CD45+ HSPCs) and their subpopulations identified and visualized as clusters using uniform manifold approximation and projection (UMAP), stressing the need for an integrated analysis of both datasets, which may be merged and treated as “pseudobulk.” We revealed that both populations do not differ significantly in gene expression, as evidenced by the very strong positive linear relationship between these cells (R = 0.99).

To obtain solid results that allow to draw conclusions that would have a biological translation, all parts of the scRNA-seq experiment are crucial and must be carried out with due care: cell sorting, single cell libraries preparation, quality control, and data analysis. The idea of working with sorted material instead of the typical use of a full pellet of blood cells was right and confirmed the possibility of HSPC analysis, even with a limited number of cells.