Long lived transients in gene regulation

• DOI: 10.1016/j.tcs.2021.05.023
• Publication Date: 2021-06-04T17:49:45Z
• Authors: Tatjana Petrov; Claudia Igler; Ali Sezgin; Thomas A. Henzinger; Calin C. Guet
• Keywords: 0301 basic medicine; 0303 health sciences; Gene regulation, Stochastic modelling, Transient memory, Long lived transients, Finite state dynamical systems, DNA looping, Rule-based modelling; Transient memory; Computer circuits; DNA; Dynamical systems; Feedback; Modeling languages; Stochastic models; Toys; Transcription; Transient analysis; Long lived transients; Gene regulation; 03 medical and health sciences; Gene regulation; Stochastic modelling; Transient memory; Long lived transients; Finite state dynamical systems; DNA looping; Rule-based modelling; DNA looping; Rule-based modelling; info:eu-repo/classification/ddc/004; Stochastic modelling; Finite state dynamical systems

Gene expression is regulated by the set of transcription factors (TFs) that bind to promoter. The ensuing regulating function often represented as a combinational logic circuit, where output (gene expression) determined current input values (promoter bound TFs) only. However, simultaneous arrival TFs strong assumption, since and translation genes introduce intrinsic time delays there no global synchronisation among times different molecular species at their targets. We present an experimentally implementable genetic circuit with two inputs one output, which in presence small arrival, exhibits qualitatively distinct population-level phenotypes, over timescales are longer than typical cell doubling times. From dynamical systems point view, these phenotypes represent long-lived transients: although they converge same value eventually, do so after very long span. key feature this toy model that, despite having only it twenty-three DNA-TF configurations, more stable others (DNA looped states), promoting another blocking gene. Small result majority cells population quickly reaching state associated first input, while exiting occurs slow timescale. In order mechanistically behaviour we used rule-based modelling language, implemented grid-search find parameter combinations giving rise transients. Our analysis shows absence feedback, exist path-dependent gene regulatory mechanisms based on timescale suggests networks can exploit event timing create opens possibility could use memorise events, without feedback. reveals importance (i) transitions between states, (ii) employing transient thereof.