Differential expression analysis in single-cell transcriptomics enables the dissection of cell-type-specific responses to perturbations such as disease, trauma, or experimental manipulations. While many statistical methods are available to identify differentially expressed genes, the principles that distinguish these methods and their performance remain unclear. Here, we show that the relative performance of these methods is contingent on their ability to account for variation between biological replicates. Methods that ignore this inevitable variation are biased and prone to false discoveries. Indeed, the most widely used methods can discover hundreds of differentially expressed genes in the absence of biological differences. To exemplify these principles, we exposed true and false discoveries of differentially expressed genes in the injured mouse spinal cord.

Differential expression analysis in single-cell transcriptomics enables the dissection of cell-type-specific responses to perturbations such as disease, trauma, or experimental manipulation. While many statistical methods are available to identify differentially expressed genes, the principles that distinguish these methods and their performance remain unclear. Here, we show that the relative performance of these methods is contingent on their ability to account for variation between biological replicates. Methods that ignore this inevitable variation are biased and prone to false discoveries. Indeed, the most widely used methods can discover hundreds of differentially expressed genes in the absence of biological differences. Our results suggest an urgent need for a paradigm shift in the methods used to perform differential expression analysis in single-cell data.

Sensorimotor integration is essential for controlling movement and acquiring new motor tasks in humans. The aim of this project was to understand how lower limb proprioceptive sense contributes to the acquisition of a skilled walking task. We assessed lower limb joint position and movement detection sense in healthy human subjects using the Lokomat robotic exoskeleton. Subjects walked on a treadmill to practice a skilled motor task (200 trials) requiring them to match their foot height during the swing phase to the height of a virtual obstacle displayed on a monitor in front of them. Subjects were given visual feedback on their error relative to the obstacle height after it was crossed. Lower limb joint position sense was related to the final performance error, but not the learning rate of the skilled walking task. The findings from this study support the role of lower limb proprioceptive sense on locomotor skill performance in healthy adult subjects.