Trapnell, C. et al. Nat. Biotechnol. 32, 381–386 (2014).

Following gene expression over time is critical to understanding dynamic cellular processes such as differentiation, and collecting data from single cells has the potential to untangle these phenomena at high resolution. Trapnell et al. introduce Monocle, an unsupervised algorithm that can place single cells in 'pseudotemporal' order on the basis of their expression profiles. The researchers used Monocle to investigate the differentiation of myoblast progenitors into muscle. They were able to identify key events in differentiation that were not detected by bulk cell sequencing or single-cell profiles ordered by time of collection. Monocle also detected alternate differentiation trajectories and a group of undifferentiated cells. The increased temporal resolution allowed the researchers to perform better gene coexpression analysis and identify eight new transcription factors implicated in muscle development.