Pooling across cells to normalize single-cell RNA sequencing data with many zero counts

TLDR

Normalization of single‑cell RNA‑seq data is essential to remove cell‑specific biases, yet the prevalence of zero counts makes this task difficult. The authors propose a novel strategy that sums expression values across pools of cells to generate normalization factors. Pool‑based size factors are computed from these summed values and subsequently deconvolved to recover cell‑specific normalization factors. The deconvolution approach outperforms existing methods in simulations and enhances the relevance of downstream analyses in real data.

Abstract

Normalization of single-cell RNA sequencing data is necessary to eliminate cell-specific biases prior to downstream analyses. However, this is not straightforward for noisy single-cell data where many counts are zero. We present a novel approach where expression values are summed across pools of cells, and the summed values are used for normalization. Pool-based size factors are then deconvolved to yield cell-based factors. Our deconvolution approach outperforms existing methods for accurate normalization of cell-specific biases in simulated data. Similar behavior is observed in real data, where deconvolution improves the relevance of results of downstream analyses.