Abstract
Analysis of single-cell transcriptomes can identify cell populations more abundant in one sample or condition than another. However, existing methods to discover them suffer from either low discovery rates or high rates of false positives. We introduce Dawnn, a deep neural network able to find differential abundance with higher accuracy than current tools, both on simulated and biological datasets. Further, we demonstrate that Dawnn recovers published findings and discovers more cells in regions of differential abundance than existing methods, both in abundant and rare cell types, promising novel biological insights at single-cell resolution.
Competing Interest Statement
The authors have declared no competing interest.
Footnotes
Added comparison against CNA (Reshef, Y.A., Rumker, L., Kang, J.B. et al. Co-varying neighborhood analysis identifies cell populations associated with phenotypes of interest from single-cell transcriptomics. Nat Biotechnol 40, 355-363 (2022).) Quantified number of additional cells identified as in regions of differential abundance by Dawnn compared to DA-seq.
