Dataset Information

MicroCellClust: mining rare and highly specific subpopulations from single-cell expression data.

ABSTRACT:

Motivation

Identifying rare subpopulations of cells is a critical step in order to extract knowledge from single-cell expression data, especially when the available data is limited and rare subpopulations only contain a few cells. In this paper, we present a data mining method to identify small subpopulations of cells that present highly specific expression profiles. This objective is formalized as a constrained optimization problem that jointly identifies a small group of cells and a corresponding subset of specific genes. The proposed method extends the max-sum submatrix problem to yield genes that are, for instance, highly expressed inside a small number of cells, but have a low expression in the remaining ones.

Results

We show through controlled experiments on scRNA-seq data that the MicroCellClust method achieves a high F1 score to identify rare sub-populations of artificially planted human T cells. The effectiveness of MicroCellClust is confirmed as it reveals a subpopulation of CD4 T cells with a specific phenotype from breast cancer samples, and a subpopulation linked to a specific stage in the cell cycle from breast cancer samples as well. Finally, 3 rare subpopulations in mouse embryonic stem cells are also identified with MicroCellClust. These results illustrate the proposed method outperforms typical alternatives at identifying small subsets of cells with highly specific expression profiles.

Availability

The R and Scala implementation of MicroCellClust is freely available on GitHub, at https://github.com/agerniers/MicroCellClust/ The data underlying this article are available on Zenodo, at https://dx.doi.org/10.5281/zenodo.4580332.

Supplementary information

Supplementary data are available at Bioinformatics online.

SUBMITTER: Gerniers A

PROVIDER: S-EPMC8504615 | biostudies-literature |

REPOSITORIES: biostudies-literature

ACCESS DATA

Similar Datasets

Project description:Objective Ovarian cancer is the deadliest gynaecological cancer globally. In our study, we aimed to analyze specific cell subpopulations and marker genes among ovarian cancer cells by single-cell RNA sequencing (RNA-seq). Methods Single-cell RNA-seq data of 66 high-grade serous ovarian cancer cells were employed from the Gene Expression Omnibus (GEO). Using the Seurat package, we performed quality control to remove cells with low quality. After normalization, we detected highly variable genes across the single cells. Then, principal component analysis (PCA) and cell clustering were performed. The marker genes in different cell clusters were detected. A total of 568 ovarian cancer samples and 8 normal ovarian samples were obtained from The Cancer Genome Atlas (TCGA) database. Differentially expressed genes were identified according to ∣log2fold change (FC) | >1 and adjusted p value <0.05. To explore potential biological processes and pathways, functional enrichment analyses were performed. Furthermore, survival analyses of differentially expressed marker genes were performed. Results After normalization, 6000 highly variable genes were identified across the single cells. The cells were divided into 3 cell populations, including G1, G2M, and S cell cycles. A total of 1,124 differentially expressed genes were identified in ovarian cancer samples. These differentially expressed genes were enriched in several pathways associated with cancer, such as metabolic pathways, pathways in cancer, and PI3K-Akt signaling pathway. Furthermore, marker genes, STAT1, ANP32E, GPRC5A, and EGFL6, were highly expressed in ovarian cancer, while PMP22, FBXO21, and CYB5R3 were lowly expressed in ovarian cancer. These marker genes were positively associated with prognosis of ovarian cancer. Conclusion Our findings revealed specific cell subpopulations and marker genes in ovarian cancer using single-cell RNA-seq, which provided a novel insight into the heterogeneity of ovarian cancer.