Data-driven learning of structure augments quantitative prediction of biological responses.

Multi-factor screenings are commonly used in diverse applications in medicine and bioengineering, including optimizing combination drug treatments and microbiome engineering. Despite the advances in high-throughput technologies, large-scale experiments typically remain prohibitively expensive. Here we introduce a machine learning platform, structure-augmented regression (SAR), that exploits the intrinsic structure of each biological system to learn a high-accuracy model with minimal data requirement. Under different environmental perturbations, each biological system exhibits a unique, structured phenotypic response. This structure can be learned based on limited data and once learned, can constrain subsequent quantitative predictions. We demonstrate that SAR requires significantly fewer data comparing to other existing machine-learning methods to achieve a high prediction accuracy, first on simulated data, then on experimental data of various systems and input dimensions. We then show how a learned structure can guide effective design of new experiments. Our approach has implications for predictive control of biological systems and an integration of machine learning prediction and experimental design.

Engineered microbial consortia: strategies and applications.

Many applications of microbial synthetic biology, such as metabolic engineering and biocomputing, are increasing in design complexity. Implementing complex tasks in single populations can be a challenge because large genetic circuits can be burdensome and difficult to optimize. To overcome these limitations, microbial consortia can be engineered to distribute complex tasks among multiple populations. Recent studies have made substantial progress in programming microbial consortia for both basic understanding and potential applications. Microbial consortia have been designed through diverse strategies, including programming mutualistic interactions, using programmed population control to prevent overgrowth of individual populations, and spatial segregation to reduce competition. Here, we highlight the role of microbial consortia in the advances of metabolic engineering, biofilm production for engineered living materials, biocomputing, and biosensing. Additionally, we discuss the challenges for future research in microbial consortia.

A cell-free biosynthesis platform for modular construction of protein glycosylation pathways.

Glycosylation plays important roles in cellular function and endows protein therapeutics with beneficial properties. However, constructing biosynthetic pathways to study and engineer precise glycan structures on proteins remains a bottleneck. Here, we report a modular, versatile cell-free platform for glycosylation pathway assembly by rapid in vitro mixing and expression (GlycoPRIME). In GlycoPRIME, glycosylation pathways are assembled by mixing-and-matching cell-free synthesized glycosyltransferases that can elaborate a glucose primer installed onto protein targets by an N-glycosyltransferase. We demonstrate GlycoPRIME by constructing 37 putative protein glycosylation pathways, creating 23 unique glycan motifs, 18 of which have not yet been synthesized on proteins. We use selected pathways to synthesize a protein vaccine candidate with an α-galactose adjuvant motif in a one-pot cell-free system and human antibody constant regions with minimal sialic acid motifs in glycoengineered Escherichia coli. We anticipate that these methods and pathways will facilitate glycoscience and make possible new glycoengineering applications.