Currently, predictive translation tuning of regulatory elements to the desired output of transcription factor (TF)-based biosensors remains a challenge. The gene expression of a biosensor system must exhibit appropriate translation intensity, which is controlled by the ribosome-binding site (RBS), to achieve fine-tuning of its dynamic range (i.e. fold change in gene expression between the presence and absence of inducer) by adjusting the translation level of the TF and reporter. However, existing TF-based biosensors generally suffer from unpredictable dynamic range. Here, we elucidated the connections and partial mechanisms between RBS, translation level, protein folding and dynamic range, and presented a design platform that predictably tuned the dynamic range of biosensors based on deep learning of large datasets cross-RBSs (cRBSs). In doing so, a library containing 7053 designed cRBSs was divided into five sub-libraries through fluorescence-activated cell sorting to establish a classification model based on convolutional neural network in deep learning. Finally, the present work exhibited a powerful platform to enable predictable translation tuning of RBS to the dynamic range of biosensors.

Promoters are one of the most critical regulatory elements controlling metabolic pathways. However, the fast and accurate prediction of promoter strength remains challenging, leading to time- and labor-consuming promoter construction and characterization processes. This dilemma is caused by the lack of a big promoter library that has gradient strengths, broad dynamic ranges, and clear sequence profiles that can be used to train an artificial intelligence model of promoter strength prediction. To overcome this challenge, we constructed and characterized a mutant library of Trc promoters (P trc) using 83 rounds of mutation-construction-screening-characterization engineering cycles. After excluding invalid mutation sites, we established a synthetic promoter library that consisted of 3665 different variants, displaying an intensity range of more than two orders of magnitude. The strongest variant was ∼69-fold stronger than the original P trc and 1.52-fold stronger than a 1 mM isopropyl-β-d-thiogalactoside-driven P T7 promoter, with an ∼454-fold difference between the strongest and weakest expression levels. Using this synthetic promoter library, different machine learning models were built and optimized to explore the relationships between promoter sequences and transcriptional strength. Finally, our XgBoost model exhibited optimal performance, and we utilized this approach to precisely predict the strength of artificially designed promoter sequences (R 2 = 0.88, mean absolute error = 0.15, and Pearson correlation coefficient = 0.94). Our work provides a powerful platform that enables the predictable tuning of promoters to achieve optimal transcriptional strength.

23Currently, predictive translation tuning of regulatory elements to the desired output of 24 transcription factor based biosensors remains a challenge. The gene expression of a biosensor 25 system must exhibit appropriate translation intensity, which is controlled by the ribosome-binding 26 site (RBS), to achieve fine-tuning of its dynamic range (i.e., fold change in gene expression between 27 the presence and absence of inducer) by adjusting the translation initiation rate of the transcription 28 factor and reporter. However, existing genetically encoded biosensors generally suffer from 29 unpredictable translation tuning of regulatory elements to dynamic range. Here, we elucidated the 30 connections and partial mechanisms between RBS, translation initiation rate, protein folding and 31 dynamic range, and presented a rational design platform that predictably tuned the dynamic range 32 of biosensors based on deep learning of large datasets cross-RBSs (cRBSs). A library containing 33 24,000 semi-rationally designed cRBSs was constructed using DNA microarray, and was divided 34into five sub-libraries through fluorescence-activated cell sorting. To explore the relationship 35 between cRBSs and dynamic range, we established a classification model with the cRBSs and 36 average dynamic range of five sub-libraries to accurately predict the dynamic range of biosensors 37 based on convolutional neural network in deep learning. Thus, this work provides a powerful 38 platform to enable predictable translation tuning of RBS to the dynamic range of biosensors. 39 40