A novel model class for bowtie biological networks with universal classification properties

TL;DR

This paper introduces a new model for resilient bacterial transcription networks, combining known motifs into a bowtie structure to analyze their capacity for environmental classification and adaptive response.

Contribution

It presents a novel hybrid network motif model (DOR2SIM) that captures the classification capabilities of resilient bacterial transcription networks.

Findings

Low monomer degradation rates facilitate classification.

Low source node gene expression at equilibrium supports classification.

The model demonstrates the network's capacity for environmental decision making.

Abstract

Cell sensory transcription networks are the intracellular computation structure that regulates and drives cellular activity. Activity in these networks determines the the cell's ability to adapt to changes in its environment. Resilient cells successfully identify (classify) and appropriately respond to environmental shifts. We present a model for identification and response to environmental changes in resilient bacteria. This model combines two known motifs in transcription networks: dense overlapping regulons (DORs) and single input modules (SIMs). DORs have the ability to perform cellular decision making and have a network structure similar to that of a shallow neural network, with a number of input transcription factors (TFs) mapping to a distinct set of genes. SIMs contain a master TF that simultaneously activates a number of target genes. Within most observed cell sensory transcription networks, the master transcription factor of SIMs are output genes of a DOR creating a fan-in-fan-out (bowtie) structure in the transcriptional network. We model this hybrid network motif (which we call the DOR2SIM motif) with a superposition of modular nonlinear functions to describe protein signaling in the network and basic mass action kinetics to describe the other chemical reactions in this process. We analyze this model's biological feasibility and capacity to perform classification, the first step in adaptation. We provide sufficient conditions for models of the DOR2SIM motif to classify constant (environmental) inputs. These conditions suggest that generally low monomer degradation rates as well as low expression of source node genes at equilibrium in the DOR component enable classification.