Neural network and kinetic modelling of human genome replication reveal replication origin locations and strengths

Abstract

In human and other metazoans, the determinants of replication origin location and strength are still elusive. Origins are licensed in G1 phase and fired in S phase of the cell cycle, respectively. It is debated which of these two temporally separate steps determines origin efficiency. Experiments can independently profile mean replication timing (MRT) and replication fork directionality (RFD) genome-wide. Such profiles contain information on multiple origins’ properties and on fork speed. Due to possible origin inactivation by passive replication, however, observed and intrinsic origin efficiencies can markedly differ. Thus, there is a need for methods to infer intrinsic from observed origin efficiency, which is context-dependent. Here, we show that MRT and RFD data are highly consistent with each other but contain information at different spatial scales. Using neural networks, we infer an origin licensing landscape that, when inserted in an appropriate simulation framework, jointly predicts MRT and RFD data with unprecedented precision and underlies the importance of dispersive origin firing. We furthermore uncover an analytical formula that predicts intrinsic from observed origin efficiency combined with MRT data. Comparison of inferred intrinsic origin efficiencies with experimental profiles of licensed origins (ORC, MCM) and actual initiation events (Bubble-seq, SNS-seq, OK-seq, ORM) show that intrinsic origin efficiency is not solely determined by licensing efficiency. Thus, human replication origin efficiency is set at both the origin licensing and firing steps.

DNA replication and polymer chain duplication reshape the genome in space and time

“…[13], in which the authors synchronized cells in G1 and then measured a time course of the 3C contacts at different times after the release. To be comparable with our predictions, we compute the normalized average Micro-C contact maps around the top 15% more efficient origins of replications [40] (see Materials and Methods) (Fig. 5A).…”

“…(Top) Progression of the replication forks in presence of putative interacting forks. (Bottom) Experimentally-determined Mean Replication Time profile (MRT) (from [40]) in the region of interest. Higher MRT values correspond to early replicating regions.…”

“…To characterize the 3D organization of chromosomes during DNA replication, we introduce a thermodynamic framework that combines a 1D stochastic description of the replication process [28,27] with a 3D polymer model of the chromosome spatio-temporal dynamics [21], including explicitly chain duplication.…”

“…To define the top 15% origins of replication, we used the "initiation probability landscape" signal (IP LS) from [40].The IP LS is a signal which defines origin efficiency and is obtained through a machine learning approach in order to accurately reproduce experimental mean replication timing and Replication Forks Directionality data [69]. We use the Scipy function f ind peak [71] to extract the peaks of IP LS .…”

“…However, to date, most theoretical approaches have focused on genome folding during the G1 phase[21, 22] and, to a lesser degree, on late G2 phase and mitosis [23, 24, 25, 26]. Furthermore, although several modelling studies have previously tackled the 1D dynamics of eukaryotic replication [27, 28, 29], a theoretical investigation of its implications on 3D structure remains largely lacking. Recent polymer models have investigated the replication process in the context of circular and highly confined bacterial chromosomes [30, 31], highlighting the role of entropic repulsion triggered by the various loops formed during replication.…”