Identifying the combinatorial control of signal-dependent transcription factors

The effectiveness of immune responses depends on the precision of stimulus-responsive gene expression programs. Cells specify which genes to express by activating stimulus-specific combinations of stimulus-induced transcription factors (TFs). Their activities are decoded by a gene regulatory strategy (GRS) associated with each response gene. Here, we examined whether the GRSs of target genes may be inferred from stimulus-response (input-output) datasets, which remains an unresolved model-identifiability challenge. We developed a mechanistic modeling framework and computational workflow to determine the identifiability of all possible combinations of synergistic (AND) or non-synergistic (OR) GRSs involving three transcription factors. Considering different sets of perturbations for stimulus-response studies, we found that two thirds of GRSs are easily distinguishable but that substantially more quantitative data is required to distinguish the remaining third. To enhance the accuracy of the inference with timecourse experimental data, we developed an advanced error model that avoids error overestimates by distinguishing between value and temporal error. Incorporating this error model into a Bayesian framework, we show that GRS models can be identified for individual genes by considering multiple datasets. Our analysis rationalizes the allocation of experimental resources by identifying most informative TF stimulation conditions. Applying this computational workflow to experimental data of immune response genes in macrophages, we found that a much greater fraction of genes are combinatorially controlled than previously reported by considering compensation among transcription factors. Specifically, we revealed that a group of known NF kappa B target genes may also be regulated by IRF3, which is supported by chromatin immuno-precipitation analysis. Our study provides a computational workflow for designing and interpreting stimulus-response gene expression studies to identify underlying gene regulatory strategies and further a mechanistic understanding. Author summary Cells need to sense environmental cues and respond appropriately. One important notion is that different stimuli activate different combinations of transcription factors and that responsive genes are regulated by distinct subsets of these. However, identifying the regulatory strategies by which genes interpret transcription factor activities remains a largely unsolved challenge. In this work we address the question: to what extent are combinatorial transcription factor regulatory strategies identifiable from stimulus-response (input-output) datasets? We present a computational framework to determine the identifiability of gene regulatory strategies, and examine how reliable and quantitative model inference is a function of the quality and quantity of available data. We present an error model that more precisely quantifies uncertainty for perturbation-timecourse data sets by also considering error in the time domain, and achieves an improved performance in identifying and quantifying gene regulatory strategies. With these tools, we generate guidelines for experimental design that optimize limited resources for generating data for model inference. Finally, we apply the workflow to immune response datasets and uncover evidence that many more genes are subject to combinatorial control that previously thought; we offer physical transcription factor binding data to support this finding for one particular group of genes. This demonstrates that the computational workflow may guide studies of the regulatory strategies that govern stimulus-responsive gene expression programs.

Automated summary

This study explores the identifiability of gene regulatory strategies using a mechanistic modeling approach and computational workflow. Results show that most GRS can be easily distinguished, but a third require more quantitative data. By developing an advanced error model and incorporating it into a Bayesian framework, GRS models for individual genes can be identified with multiple datasets, allowing for better allocation of experimental resources. Applying this workflow to immune response genes in macrophages revealed more combinatorial control than previously thought, supported by chromatin immuno-precipitation analysis.