Identifying Cell Types in Single-Cell RNA Sequencing Data
Author Information
Author(s): Nasrollahi Fatemeh Sadat Fatemi, Silva Filipi Nascimento, Liu Shiwei, Chaudhuri Soumilee, Yu Meichen, Wang Juexin, Nho Kwangsik, Saykin Andrew J., Bennett David A., Sporns Olaf, Fortunato Santo
Primary Institution: Indiana University
Hypothesis
Can network clustering algorithms improve the identification of cell types in single-cell RNA sequencing data?
Conclusion
The study found that the Infomap algorithm is particularly effective for accurately identifying cell types in single-cell RNA sequencing data.
Supporting Evidence
- Infomap showed a closer alignment with actual cell types compared to other methods.
- Clustering algorithms were tested on two independent datasets.
- Preprocessing steps significantly influenced clustering performance.
Takeaway
This study helps scientists figure out what different types of cells are in a sample by using smart computer methods to look at tiny bits of information from each cell.
Methodology
The study used various clustering algorithms on cell-cell networks derived from gene expression data from two datasets, PBMC and ROSMAP.
Limitations
The study assumes that algorithmically detected clusters align perfectly with predefined cell types, which may not always be the case.
Participant Demographics
The ROSMAP dataset included 12 male and 12 female participants, while the PBMC dataset comprised a heterogeneous population of white blood cells.
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website