DyMMM-LEAPS: An ML-based framework for modulating evenness and stability in synthetic microbial communities.

There have been a growing number of computational strategies to aid in the design of synthetic microbial consortia. A framework to identify regions in parametric space to maximize two essential properties, evenness and stability, is critical. In this study, we introduce DyMMM-LEAPS (dynamic multispecies metabolic modeling-locating evenness and stability in large parametric space), an extension of the DyMMM framework. Our method explores the large parametric space of genetic circuits in synthetic microbial communities to identify regions of evenness and stability. Due to the high computational costs of exhaustive sampling, we utilize adaptive sampling and surrogate modeling to reduce the number of simulations required to map the vast space. Our framework predicts engineering targets and computes their operating ranges to maximize the probability of the engineered community to have high evenness and stability. We demonstrate our approach by simulating five cocultures and one three-strain culture with different social interactions (cooperation, competition, and predation) employing quorum-sensing-based genetic circuits. In addition to guiding circuit tuning, our pipeline gives an opportunity for a detailed analysis of pockets of evenness and stability for the circuit under investigation, which can further help dissect the relationship between the two properties. DyMMM-LEAPS is easily customizable and can be expanded to a larger community with more complex interactions.

Engineering a short, aldolase-based pathway for (R)-1,3-butanediol production in Escherichia coli.

Microbial processes can produce a wide range of compounds; however, producing complex and long chain hydrocarbons remains a challenge. Aldol condensation offers a direct route to synthesize these challenging chemistries and can be catalyzed by microbes using aldolases. Deoxyribose-5-phosphate aldolase (DERA) condenses aldehydes and/or ketones to ß-hydroxyaldehydes, which can be further converted to value-added chemicals such as a precursor to cholesterol-lowering drugs. Here, we implement a short, aldolase-based pathway in Escherichia coli to produce (R)-1,3-BDO from glucose, an essential component of pharmaceutical products and cosmetics. First, we expressed a three step heterologous pathway from pyruvate to produce 0.3 g/L of (R)-1,3-BDO with a yield of 11.2 mg/g of glucose in wild-type E. coli K12 MG1655. We used a systems metabolic engineering approach to improve (R)-1,3-BDO titer and yield by: 1) identifying and reducing major by-products: ethanol, acetoin, and 2,3-butanediol; 2) increasing pathway flux through DERA to reduce accumulation of toxic acetaldehyde. We then implemented a two-stage fermentation process to improve (R)-1,3-BDO titer by 8-fold to 2.4 g/L and yield by 5-fold to 56 mg/g of glucose (11% of maximum theoretical yield) in strain BD24, by controlling pH to 7 and higher dissolved oxygen level. Furthermore, this study highlights the potential of the aldolase chemistry to synthesize diverse products directly from renewable resources in microbes.

FOCUS on NOD2: Advancing IBD Drug Discovery with a User-Informed Machine Learning Framework.

In this study, we introduce the Framework for Optimized Customizable User-Informed Synthesis (FOCUS), a generative machine learning model tailored for drug discovery. FOCUS integrates domain expertise and uses Proximal Policy Optimization (PPO) to guide Monte Carlo Tree Search (MCTS) to efficiently explore chemical space. It generates SMILES representations of potential drug candidates, optimizing for druggability and binding efficacy to NOD2, PEP, and MCT1 receptors. The model is highly interpretive, allowing for user-feedback and expert-driven adjustments based on detailed cycle reports. Employing tools like SHAP and LIME, FOCUS provides a transparent analysis of decision-making processes, emphasizing features such as docking scores and interaction fingerprints. Comparative studies with Muramyl Dipeptide (MDP) demonstrate improved interaction profiles. FOCUS merges advanced machine learning with expert insight, accelerating the drug discovery pipeline.

Toward a systematic design of smart probiotics.

Synthetic biology has been instrumental in turning microbes into cell-factories capable of diverse processes. A recent application has been to convert them into living therapeutics with diagnostic and production capabilities. These smart probiotics act as living medicines inside the body capable of diagnosing and responding to environmental cues in real time. Many companies and research groups have reported success with smart probiotics with several advancing to human clinical trials. Despite the promise, engineered probiotics are unable to replicate their functionality in a more complex environment. A rich environment, such as the gut, imposes restrictions on probiotic durability and effectiveness, hindering its ability to reach its full potential. Scientists have a plethora of advanced tools available today that enable enhanced strain engineering decisions to increase the production of fuels and commodity chemicals. However, these tools have not yet found mainstream application in building smart probiotics. Majority of the work in this field still relies on rational engineering. This review will propose strategies, with a focus on model-based approaches, that can help bridge the gap to systematic design and optimization of smart probiotics.