An arginine-rich nuclear localization signal (ArgiNLS) strategy for streamlined image segmentation of single-cells.

Quantifying labeled cells in fluorescent microscopy is a fundamental aspect of a modern biology. Critically, the use of short nuclear localization sequences (NLS) is a key genetic modification for discriminating single-cells labeled with fluorescent proteins (FPs). However, mainstay NLS approaches typically localize proteins to the nucleus with limited efficacy, while alternative non-NLS tag strategies can enhance efficacy at the cost of cellular health. Thus, quantitative cell counting using FP labels remains suboptimal or not compatible with health and behavior. Here, we present a novel genetic tagging strategy - named ArgiNLS - that flexibly and safely achieves FP nuclear restriction across the brain to facilitate machine learning-based segmentation of single-cells at scale, delivering a timely update to the behavioral neuroscientist's toolkit.