Comparison and evaluation of statistical error models for scRNA-seq

BACKGROUND

Heterogeneity in single-cell RNA-seq (scRNA-seq) data is driven by multiple sources, including biological variation in cellular state as well as technical variation introduced during experimental processing. Deconvolving these effects is a key challenge for preprocessing workflows. Recent work has demonstrated the importance and utility of count models for scRNA-seq analysis, but there is a lack of consensus on which statistical distributions and parameter settings are appropriate.

RESULTS

Here, we analyze 59 scRNA-seq datasets that span a wide range of technologies, systems, and sequencing depths in order to evaluate the performance of different error models. We find that while a Poisson error model appears appropriate for sparse datasets, we observe clear evidence of overdispersion for genes with sufficient sequencing depth in all biological systems, necessitating the use of a negative binomial model. Moreover, we find that the degree of overdispersion varies widely across datasets, systems, and gene abundances, and argues for a data-driven approach for parameter estimation.

CONCLUSIONS

Based on these analyses, we provide a set of recommendations for modeling variation in scRNA-seq data, particularly when using generalized linear models or likelihood-based approaches for preprocessing and downstream analysis.

Modified horseshoe crab peptides target and kill bacteria inside host cells

Bacteria that occupy an intracellular niche can evade extracellular host immune responses and antimicrobial molecules. In addition to classic intracellular pathogens, other bacteria including uropathogenic Escherichia coli (UPEC) can adopt both extracellular and intracellular lifestyles. UPEC intracellular survival and replication complicates treatment, as many therapeutic molecules do not effectively reach all components of the infection cycle. In this study, we explored cell-penetrating antimicrobial peptides from distinct structural classes as alternative molecules for targeting bacteria. We identified two β-hairpin peptides from the horseshoe crab, tachyplesin I and polyphemusin I, with broad antimicrobial activity toward a panel of pathogenic and non-pathogenic bacteria in planktonic form. Peptide analogs [I11A]tachyplesin I and [I11S]tachyplesin I maintained activity toward bacteria, but were less toxic to mammalian cells than native tachyplesin I. This important increase in therapeutic window allowed treatment with higher concentrations of [I11A]tachyplesin I and [I11S]tachyplesin I, to significantly reduce intramacrophage survival of UPEC in an in vitro infection model. Mechanistic studies using bacterial cells, model membranes and cell membrane extracts, suggest that tachyplesin I and polyphemusin I peptides kill UPEC by selectively binding and disrupting bacterial cell membranes. Moreover, treatment of UPEC with sublethal peptide concentrations increased zinc toxicity and enhanced innate macrophage antimicrobial pathways. In summary, our combined data show that cell-penetrating peptides are attractive alternatives to traditional small molecule antibiotics for treating UPEC infection, and that optimization of native peptide sequences can deliver effective antimicrobials for targeting bacteria in extracellular and intracellular environments.