An official website of the United States government
Editorial overview: Not everyone can become a cell biologist, but a great cell biologist can come from anywhere
- PAR ID:
- 10439145
- Publisher / Repository:
- Elsevier
- Date Published:
- Journal Name:
- Current Opinion in Plant Biology
- Volume:
- 73
- Issue:
- C
- ISSN:
- 1369-5266
- Page Range / eLocation ID:
- 102367
- Format(s):
- Medium: X
- Sponsoring Org:
- National Science Foundation
More Like this
-
Synopsis Single-cell RNA sequencing (scRNAseq) is a powerful tool to describe cell types in multicellular organisms across the animal kingdom. In standard scRNAseq analysis pipelines, clusters of cells with similar transcriptional signatures are given cell type labels based on marker genes that infer specialized known characteristics. Since these analyses are designed for model organisms, such as humans and mice, problems arise when attempting to label cell types of distantly related, non-model species that have unique or divergent cell types. Consequently, this leads to limited discovery of novel species-specific cell types and potential mis-annotation of cell types in non-model species while using scRNAseq. To address this problem, we discuss recently published approaches that help annotate scRNAseq clusters for any non-model organism. We first suggest that annotating with an evolutionary context of cell lineages will aid in the discovery of novel cell types and provide a marker-free approach to compare cell types across distantly related species. Secondly, machine learning has greatly improved bioinformatic analyses, so we highlight some open-source programs that use reference-free approaches to annotate cell clusters. Lastly, we propose the use of unannotated genes as potential cell markers for non-model organisms, as many do not have fully annotated genomes and these data are often disregarded. Improving single-cell annotations will aid the discovery of novel cell types and enhance our understanding of non-model organisms at a cellular level. By unifying approaches to annotate cell types in non-model organisms, we can increase the confidence of cell annotation label transfer and the flexibility to discover novel cell types.more » « less
-
Weaver, Valerie Marie (Ed.)Chromosome numbers often change dynamically in tumors and cultured cells, which complicates therapy as well as understanding genotype-mechanotype relationships. Here we use a live-cell “ChReporter” method to identify cells with a single chromosomal loss in efforts to better understand differences in cell shape, motility, and growth. We focus on a standard cancer line and first show clonal populations that retain the ChReporter exhibit large differences in cell and nuclear morphology as well as motility. Phenotype metrics follow simple rules, including migratory persistence scaling with speed, and cytoskeletal differences are evident from drug responses, imaging, and single-cell RNA sequencing. However, mechanotype–genotype relationships between fluorescent ChReporter-positive clones proved complex and motivated comparisons of clones that differ only in loss or retention of a Chromosome-5 ChReporter. When lost, fluorescence-null cells show low expression of Chromosome-5 genes, including a key tumor suppressor APC that regulates microtubules and proliferation. Colonies are compact, nuclei are rounded, and cells proliferate more, with drug results implicating APC, and patient survival data indicating an association in multiple tumor-types. Visual identification of genotype with ChReporters can thus help clarify mechanotype and mechano-evolution.more » « less
