Single-cell and spatial transcriptomics clustering with an optimized adaptive k-nearest neighbor graph
Presenting author: Jia Li, Department of Biostatistics, Vanderbilt University Medical Center
Co-authored by:
- Yu Shyr, Department of Biostatistics, Vanderbilt University Medical Center
- Qi Liu, Department of Biostatistics, Vanderbilt University Medical Center
Abstract:
Typical clustering methods for single-cell and spatial transcriptomics have difficulty in identifying rare cell types, while approaches specifically tailored to detect rare cell types gain their ability at the cost of poorer performance for grouping abundant ones. Here, we developed aKNNO to identify abundant and rare cell types simultaneously based on an adaptive k-nearest neighbor graph with optimization. Benchmarked on 38 simulated and 20 single-cell and spatial transcriptomics datasets, aKNNO identified both abundant and rare cell types more accurately than those typical and specifically tailored methods. aKNNO, using transcriptome alone, stereotyped fine-grained anatomical structures more precisely than those integrative approaches.