Precision engineering of biological function with large-scale measurements and machine learning

• DOI: 10.1371/journal.pone.0283548
• Publication Date: 2023-03-29T17:56:17Z
• Authors: Drew S. Tack; Peter D. Tonner; Abe Pressman; Nathan D. Olson; Sasha F. Levy; Eugenia F. Romant...; Nina Alperovich; Olga Vasilyeva; David Ross
• Keywords: Machine Learning; 0301 basic medicine; 03 medical and health sciences; Science; Q; R; Medicine; Synthetic Biology; Research Article

As synthetic biology expands and accelerates into real-world applications, methods for quantitatively precisely engineering biological function become increasingly relevant. This is particularly true applications that require programmed sensing to dynamically regulate gene expression in response stimuli. However, few have been described can engineer with any level of quantitative precision. Here, we present two complementary precision genetic sensors: silico selection machine-learning-enabled forward engineering. Both use a large-scale genotype-phenotype dataset identify DNA sequences encode sensors specified dose response. First, show be used wide range dose-response curves. To demonstrate precise, multi-objective engineering, simultaneously tune sensor’s sensitivity ( EC 50 ) saturating output meet specifications. In addition, inverted . Second, approach predictively mutation combinations are not the dataset. We interpretable machine learning results combined biophysical model improved