Protocol to analyze immune cells in the tumor microenvironment by transcriptome using machine learning.

Immunotherapy is a promising strategy to treat cancer. Here, we present a protocol for analyzing the transcriptome-based phenotypic alterations and immune cell infiltration in the tumor microenvironment. We describe steps for integrating single-cell RNA sequencing (scRNA-seq) data, comparing phenotypes and origins of mononuclear phagocytes, inferring the differentiation trajectory and infiltration process, and identifying infiltration-associated genes using machine learning. We then detail procedures for exploring the impact of these genes in prognosis through the integrated microarray and bulk RNA-seq data to obtain potential drug targets. For complete details on the use and execution of this protocol, please refer to Liao et al.1.

The efficient generation of functional human hepatocytes from chemically induced pluripotent stem cells.

Derivation of human hepatocytes from pluripotent stem cells in vitro has important applications including cell therapy and drug discovery. However, the differentiation of pluripotent stem cells into hepatocytes in vitro was not well recapitulated the development of liver. Here, we developed a differentiation protocol by mimicking the two-stage development of hepatoblasts, which permits the efficient generation of hepatic progenitor cells from chemically induced pluripotent stem cells (hCiPSCs). Single-cell RNA sequencing (scRNA-seq) indicates the similarity between hepatoblasts differentiated in vitro and in vivo. Moreover, hCiPSC-derived hepatic progenitor cells can further differentiate into hepatocytes that are similar to primary human hepatocytes with respect to gene expression and key hepatic functions. Our results demonstrate the feasibility of generating hepatic progenitor cells and hepatocytes from hCiPSCs with high efficiency and set the foundation for broad translational applications of hCiPSC-derived hepatocytes.