An international joint research group including Special Temporary Research Scientist Satoshi Oota and Team Director Hideo Yokota of the RIKEN Center for Advanced Photonics developed a new analytical method that reconstructs cell lineages and cellular diversity within normal tissues, using somatic mutations estimated from single-cell RNA sequencing (scRNA-seq) data. These lineages can be regarded as the "evolutionary records" of cells. It is expected to serve as a new analytical platform for advancing understanding of normal tissue formation and cell fate determination in developmental biology, regenerative medicine, aging research, and other fields. The results were published in DNA Research.
When cells divide, small mutations occur during DNA replication, and these accumulate in cells as somatic mutations. Somatic mutations that arise in a particular cell are inherited by its descendant cells, making them useful markers that record the history of cell divisions. Therefore, by comparing somatic mutations among cells, it is possible to estimate lineage relationships and reconstruct cell lineages.
The international research group therefore reconstructed cell lineages and cellular diversity from somatic mutations in normal cells contained in scRNA-seq data. First, they obtained RNA sequences from normal cells isolated from tissues at the single-cell level using scRNA-seq and detected somatic mutations present in individual cells based on nucleotide differences in the extracted RNA sequences. Based on differences in somatic mutation patterns, they calculated genetic distances (molecular evolutionary distances) among cells.
By applying phylogenetic tree reconstruction methods widely used in molecular phylogenetics to these results, they reconstructed cell lineages. Furthermore, through extensive simulations, they evaluated the effects of factors such as the number of somatic mutations, the number of cells, allelic dropout, and missing data characteristic of RNA sequencing on the accuracy of cell-lineage reconstruction.
As a result, they showed that cell lineages can be reconstructed with high accuracy when a sufficient number of somatic mutations are available. In addition, although human placental cells contain only somatic mutations accumulated during the relatively short period of approximately 10 months from fertilization to birth, phylogenetic branching structures corresponding to cell types (classifications based on cellular function or morphology) could be reconstructed from actual placental scRNA-seq data. Furthermore, by integrating the reconstructed cell lineages with gene expression information, they confirmed that lineage relationships and cellular diversity can be analyzed simultaneously.
The developed analytical method uses somatic mutations that naturally accumulate in normal cells as the cells' "evolutionary records," enabling direct estimation of cell lineages that represent ancestor-descendant relationships among cells. Unlike conventional approaches that infer differentiation pathways from similarities in gene expression patterns, this method utilizes evolutionary history itself, allowing cell differentiation processes to be analyzed more directly. In addition, because cell lineages can be estimated using existing scRNA-seq datasets, the method can be applied to large amounts of publicly available data without requiring new experiments.
In the future, it is expected to serve as a new analytical platform for advancing understanding of normal tissue formation and cell fate determination in areas such as developmental biology, age-related cellular changes in aging research, tissue regeneration and stem cell research in regenerative medicine, and cancer evolution studies. In addition, because cellular diversity and cell lineages can be analyzed and visualized in a unified manner, the method is expected to provide a new analytical framework for understanding how normal tissues are formed and maintained.
Journal Information
Publication: DNA Research
Title: Reconstruction of cell diversity and cell lineages from somatic mutations in single-cell transcriptomic data
DOI: 10.1093/dnares/dsag007
This article has been translated by JST with permission from The Science News Ltd. (https://sci-news.co.jp/). Unauthorized reproduction of the article and photographs is prohibited.