Abstract
Replication timing (RT) is the temporal program of genome duplication that is coordinated with 3D genome organization and gene expression(1,2). How and when RT is established during mammalian development are questions that remain unsolved. Four recent studies provided the first insights; however, they reached to contradictory conclusions. Nakatani et al. reported a poorly defined RT in mouse zygotes that is progressively consolidated by the 4-cell stage(3); Takahashi et al. and Halliwell, et al. reported a complete RT absence in the zygote with abrupt emergence at later stages(4,5); finally, Xu et al.(6) reported a RT already established in the zygote. Discrepancies might reflect differences in computational methods and thresholds applied to sparse single-cell RT data. More recently, Shetty et al. concluded that RT is established in the zygote, and that transcriptional activation and chromatin accessibility emerge at late replicating regions(7). Their conclusions challenge the long-standing relationship between gene expression and early replication(1,2,8-10) and contradict previous studies(3-6). This critique demonstrates that methodological flaws and artifacts, stemming from inadequate sample acquisition and data processing led to unreliable RT inference, rendering the conclusions on RT establishment and its inverse correlation with gene expression erroneous. Moreover, re-analysis of their data, with stage-matched controls from previous studies(3,5), demonstrates that the canonical relationship between RT, transcriptional activity, and chromatin accessibility are indeed still standing during the early mammalian embryogenesis.