Massively Parallel Experiments to Develop a Predictive Biophysical Model of Transcription Rate across Cellular Conditions

This project seeks to develop new models that quantitatively predict how bacterial gene expression is regulated across cellular and environmental conditions. The developed models enable researchers to rationally engineer microbial organisms with new sensing and metabolic capabilities that are suitable targeted environments. In bacteria, the first step in gene expression, transcription, is catalyzed by an RNA polymerase enzyme and a sigma factor protein. Bacteria contain multiple sigma factors with distinct DNA binding properties and use signaling pathways to modify the abundances of these sigma factors in response to cell state, causing large state-dependent changes in transcription rate across hundreds of genes. In this project, thousands of experiments are conducted to systematically measure how DNA nucleotide sequences control sigma factor-specific transcriptional initiation frequencies and the sites at which transcription is initiated. These measurements are used to create biophysical models of transcription initiation that accept arbitrary DNA sequence inputs, calculate the strengths of the relevant interactions, and then predict the sigma-specific frequencies of transcription initiation at each potential start site. These predictions enable the automated design of engineered genetic systems (sensors, genetic circuits, and metabolic pathways) with targeted transcriptional profiles that adapt and respond to changing cellular and environmental conditions. The biophysical models of transcription will also be combined with a web-based interface, visual animations, and tutorials to facilitate interactive and experiential student learning. This project also provides opportunities for underrepresented undergraduate students to participate in laboratory research.

This project integrates systematic design, massively parallel experiments, next-generation sequencing, and machine learning to develop predictive statistical thermodynamic models of bacterial transcription initiation (a Promoter Calculator). Transcription initiation rate measurements and transcriptional start site mapping is conducted on thousands of systematically designed promoter sequences in well-defined in vitro assays that each contain RNA polymerase and a single sigma factor. From multiple sets of measurements and using different sigma factors, sigma-specific models are trained and validated to accurately predict sigma-specific and site-specific transcription initiation rates. Model predictions are combined and validated in several ways, including by comparison to in vivo transcriptomic measurements across a range of environmental conditions and by engineering genetic systems with rationally designed state-dependent transcriptional profiles. New types of in vitro assays using naked genome templates are also devised to enhance precision in measurement and to ultimately understand transcriptional interactions at natural promoters. The developed model of transcriptional initiation will enable Synthetic Biologists to predict and control transcription rates for diverse biotech applications (e.g. engineering biosensors, genetic circuits, metabolic pathways, and genomes), while facilitating system-wide, quantitative analysis and debugging of genetic system function.

This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.