Despite the growing number of chemicals successfully engineered in host organisms, bioproduction R&D is slow and expensive, as the process is mostly based on trial-and-error. To overcome this critical hindrance, we propose to implement a generic automated design-build-test and learn cyclic pipeline for the production of targeted chemicals. As an illustration, we will apply the pipeline for the metabolic engineering of a library of new antimicrobials against Gram-positive bacteria. The pipeline comprises state-of-the-art bioproduction pathway design tools, robotized strain engineering, and high throughput product quantification via biosensors. The whole process is driven by an original computational machine learning component that determines the next set of constructions that needs to be processed by the pipeline with the goal of increasing product yield. In the specific approach we will be using, named active learning, a growing training set of experimental results is acquired on the fly in an iterative process between learning and measurements. The remarkable advantage of active learning is to yield performances comparable to classical machine learning with training sets sizes that can be several orders of magnitude smaller. Active learning can thus drastically reduce the cost of performing measurements, and in the present application significantly reduce the number of iterations for strain optimization. We propose to apply the pipeline for the production of nutritional and antimicrobial flavonoids. Precisely, the pipeline will be run for four research objectives that complement each other: (RO1) to learn enzyme sequences that maximize flavonoid titers, (RO2) to determine enzyme expression levels limiting intermediates accumulation and increasing final product yields, (RO3) to regulate the expression of the genes of the host strain to optimize both growth and flavonoid titers, and (RO4) to produce novel flavonoid structures with maximal toxicity against Gram-positive bacteria. While moving toward optimizing strains and producing novel flavonoids, our project will offer a technological rupture to industrial biotechnology where machine learning is driving experimental implementation and measurement. We anticipate this innovative solution will bring tremendous gains in throughput and speed. The project will be illustrated with the production of a library of flavonoids, but the design-build-test-learn pipeline is general enough to be applied to other molecules of interest to the health, food, chemistry and energy industrial sectors, including commodity chemicals, and fine and specialty chemicals. Our approach could for instance be extended to other pharmaceutical applications beyond the search for antimicrobial activity, as long as there exists a screening method relevant to the problem. Beyond small molecule bioproduction a similar pipeline could also be implemented to metabolize alternative but commercially attractive feedstock and to develop biosensors for environmental pollutants. The expertise gained in the project will drastically improve our SME partner strain development platform and in return the SME partner will bring the technology to the market seeking for industrial collaborations through a specific exploitation task. While we plan to release our computational methods to the academic community through web services, for specific applications, our know-how and software products will be packaged in an integrated pipeline and commercialized as a service. We foresee large industrial groups will want to customize development of the pipeline for their own application. The service we will provide to the industry will generate revenues and will also be a source for job creation.

One of the unsolved challenges of synthetic and systems biology is bottom-up prediction of gene expression from molecular interactions of regulatory molecules with the genome. Current high-throughput experiments allow generating large data sets for retrieving the DNA sequence preferences of transcription factors (TF) and other chromosomal proteins throughout the complete genome, as well as the maps of DNA methylation, histone variant incorporation and histone modifications. Due to the high combinatorial complexity of possible chromatin states an experimental determination of genome-wide maps for all possible sets of epigenetic parameters is not possible. Thus, quantitative descriptions are needed that relate molecular binding events to epigenetic states as determined from a limited number of key features. In the project proposed here, we set out to develop such a framework to predict and manipulate changes in gene expression caused by transcription factor and nucleosome rearrangements, starting from a known set of epigenetic marks that define a given cell state. To parameterize the models, we will use high-throughput sequencing data for mouse embryonic stem cell development. To validate the models, we will study the effects of novel epigenetic modulators in the same system. The resulting quantitative description will connect molecular binding events to the changes in epigenetic states at functional genomic elements during stem cell differentiation. The theoretico-experimental framework developed here will have general implications for in silico drug testing and construction of synthetic cis-regulatory modules with defined properties.

A primary goal of Synthetic Biology is to engineer or re-engineer or add additional new functionalities to a biological system by constructing new parts, or modifying an existing biological system. In White Biotechnology, a clear objective is to enrich the metabolic repertoire of a microbial cell with new metabolic enzymes or pathways that enable the production of chemicals from renewable carbon resource as an alternative cost-effective replacement of chemicals derived from fossil resources. In brief, the creation a bio-based chemical process for a given product follows a three step procedure, namely (i) pathway construction, (ii) engineering and optimization of the microbial cell factory and (iii) development of industrially-oriented process fermentation. Partner 1 of this consortium recently innovated in constructing a complex synthetic pathway composed of 8 reactions steps that leads to the production of two high added-value synthons, 2,4 dihydroxybutyrate (DHB) and 1,3 propanediol (PDO) from malate, the latter being produced from glucose using classical (natural) pathways. In this pathway, 5 reactions are catalyzed by non-natural enzymes, and three of them exhibit very poor catalytic activity and specificity towards their non-natural substrates, which renders the pathway poorly efficient. In addition, the synthetic pathway is expressed on plasmids as synthetic operons made of 3 to 8 genes. Although using plasmids, it is simple to manipulate gene expression through copy number or promoter strength, plasmids suffer from genetic instability and usually cause growth inhibition due to the ‘protein burden' resulting from their overexpression, which makes them poorly adapted for industrial applications, and raises fundamental questions about the way these synthetic operons can be to stably integrate into the host cell’s genome to eventually become a full constituent of natural cellular physiology and satisfy at the same time the ‘non-natural’ objectives. The purpose of SYNPATHIC is to provide solutions to these two major bottlenecks, enabling major advances in the field of Synthetic Biology by (i) optimizing the catalytic efficiency of individual steps in pathways, (ii) optimizing the flux through whole pathways, and (iii) simultaneously adapting the microbial cell factory to a novel ‘non-natural’ objective. These solutions will be exemplified by optimizing the pathways for DHB and PDO production from glucose. The first scientific and technical hurdle will be tackled using innovative ultrahigh-throughput screening systems for directed evolution of ‘rate-limiting’ enzymes in pathways using droplet-based microfluidics. This technology, pioneered by Partner 2, uses monodisperse aqueous droplets in a continuous oil phase as independent bio-compatible microreactors for ultrahigh-throughput biological assays. Microorganisms/genes can be screened with 1,000-fold increase in speed and a 1,000,000-fold reduction in volume (and cost) compared to screening in microtitre plates. The second bottleneck, due to genetic instability of plasmid-borne synthetic operons, will be tackled through clever genomic integration of the synthetic module. This integration will be guided by a genome design software developed by partner 3 of this consortium, based on a threefold interdependence of genome layout, DNA 3-D conformation, and co-regulation of genes functioning in common metabolic pathways. Finally, process optimization will benefit from the combination of genomic integration and droplet-based microfluidic to monitor production variability within the genome-reengineered microbial population and select for optimized producers.