A high-resolution pan-cancer T cell atlas, built using single-cell RNA sequencing data, is presented in Science this week. Tumor-specific T cells are key targets in cancer immunotherapy, but clinical efficacy varies between different types of cancer with T cell exhaustion, the antitumor functions of effector T cells, and the states and abundances of T cells across different tumor microenvironments potentially all playing a role. Aiming to better understand the heterogeneity and dynamics of tumor-infiltrating T cells across cancer types, investigators from Peking University performed single-cell RNA-seq on tumors, paracancerous tissues, and blood samples from patients of various cancer types. They combined these data with single-cell RNA-seq datasets in the literature into a single resource spanning 316 patients across 21 cancer types. The atlas, the researchers write, reveals distinct T cell composition patterns in different tumor microenvironments, multiple state-transition paths in T cell exhaustion, and the preference of these paths among different tumor types. "Our detailed signature, dynamics, and regulations of tumor-infiltrating T cells will facilitate the development of immunotherapies, and our proposed immune typing can aid the therapeutic and diagnostic strategies that target T cells," they write.
A short-read mapping tool that can accurately and rapidly map reads to thousands of genomes embedded in a pangenome reference is described in Science this week. Since a single reference genome cannot sufficiently capture the diversity within even a single person, pangenomes — which encode information about many complete genome assemblies and their homologies — have emerged as a replacement. Yet existing technologies are not efficient enough to make mapping the short sequencing reads from widely used and inexpensive DNA sequencers to a structurally complex pangenome a practical option for large-scale applications. To address this, a team led by scientists from the University of California, Santa Cruz, Genomics Institute developed Giraffe, which accurately maps sequencing reads to thousands of human genomes at a speed comparable to that of standard methods mapping to a single reference genome using a variety of algorithmic techniques. With the tool, the researchers genotyped 167,000 structural variations, recently discovered in long-read studies, in short-read samples for 5,202 people at an average computational cost of $1.50 per sample. "A single reference genome must choose one version of any variation to represent, leaving the other versions unrepresented," they write, "By making more broadly representative pangenome references practical, Giraffe attempts to make genomics more inclusive."