Adaptable control policies for variable liquid chromatography columns using deep reinforcement learning.

Controlling chromatography systems for downstream processing of biotherapeutics is challenging because of the highly nonlinear behavior of feed components and complex interactions with binding phases. This challenge is exacerbated by the highly variable binding properties of the chromatography columns. Furthermore, the inability to collect information inside chromatography columns makes real-time control even more problematic. Typical static control policies either perform sub optimally on average owing to column variability or need to be adapted for each column requiring expensive experimentation. Exploiting the recent advances in simulation-based data generation and deep reinforcement learning, we present an adaptable control policy that is learned in a data-driven manner. Our controller learns a control policy by directly manipulating the inlet and outlet flow rates to optimize a reward function that specifies the desired outcome. Training our controller on columns with high variability enables us to create a single policy that adapts to multiple variable columns. Moreover, we show that our learned policy achieves higher productivity, albeit with a somewhat lower purity, than a human-designed benchmark policy. Our study shows that deep reinforcement learning offers a promising route to develop adaptable control policies for more efficient liquid chromatography processing.

Model-based intensification of CHO cell cultures: One-step strategy from fed-batch to perfusion.

There is a growing interest in continuous processing of the biopharmaceutical industry. However, the technology transfer from traditional batch-based processes is considered a challenge as protocol and tools still remain to be established for their usage at the manufacturing scale. Here, we present a model-based approach to design optimized perfusion cultures of Chinese Hamster Ovary cells using only the knowledge captured during small-scale fed-batch experiments. The novelty of the proposed model lies in the simplicity of its structure. Thanks to the introduction of a new catch-all variable representing a bulk of by-products secreted by the cells during their cultivation, the model was able to successfully predict cellular behavior under different operating modes without changes in its formalism. To our knowledge, this is the first experimentally validated model capable, with a single set of parameters, to capture culture dynamic under different operating modes and at different scales.

High-Throughput Synthesis, Analysis, and Optimization of Injectable Hydrogels for Protein Delivery.

The development of in situ-gelling hydrogels that can enable prolonged protein release is increasingly important due to the emergence of a growing number of protein-based therapeutics. Herein, we describe a high-throughput strategy to fabricate, characterize, and subsequently optimize hydrazone-cross-linked in situ-gelling hydrogels for protein delivery. Hydrogels are fabricated using an automated high-throughput robot to mix a variety of thermoresponsive, nonthermoresponsive, charged, neutral, naturally sourced, and synthetic polymers functionalized with hydrazide or aldehyde groups, generating in situ-gelling hydrogels with well-defined compositions within a 96-well plate. High-throughput characterization strategies are subsequently developed to enable on-plate analysis of hydrogel swelling, mechanics, degradation, transparency, and protein (ovalbumin) release kinetics that yield results consistent with those collected using traditional bulk hydrogel analysis techniques. Dynamic regression and latent variable modeling are then applied to fit performance statistics to the collected data set; subsequently, numerical optimization is used to identify mixtures of precursor polymers that exhibit targeted combinations of minimal burst release, maximum total protein release, minimum release rate, and maximum transparency (the latter of particular relevance for ophthalmic protein delivery applications). Given the rapid throughput of the protocols developed (i.e., 126 hydrogels can be synthesized and screened in quadruplicate within hours), this approach offers particular promise for accelerating the identification of injectable hydrogel compositions relevant for both protein delivery as well as other biomedical applications for which clearly predefined materials properties are required.

Injectable Anisotropic Nanocomposite Hydrogels Direct in Situ Growth and Alignment of Myotubes.

While injectable in situ cross-linking hydrogels have attracted increasing attention as minimally invasive tissue scaffolds and controlled delivery systems, their inherently disorganized and isotropic network structure limits their utility in engineering oriented biological tissues. Traditional methods to prepare anisotropic hydrogels are not easily translatable to injectable systems given the need for external equipment to direct anisotropic gel fabrication and/or the required use of temperatures or solvents incompatible with biological systems. Herein, we report a new class of injectable nanocomposite hydrogels based on hydrazone cross-linked poly(oligoethylene glycol methacrylate) and magnetically aligned cellulose nanocrystals (CNCs) capable of encapsulating skeletal muscle myoblasts and promoting their differentiation into highly oriented myotubes in situ. CNC alignment occurs on the same time scale as network gelation and remains fixed after the removal of the magnetic field, enabling concurrent CNC orientation and hydrogel injection. The aligned hydrogels show mechanical and swelling profiles that can be rationally modulated by the degree of CNC alignment and can direct myotube alignment both in two- and three-dimensions following coinjection of the myoblasts with the gel precursor components. As such, these hydrogels represent a critical advancement in anisotropic biomimetic scaffolds that can be generated noninvasively in vivo following simple injection.

The effects of root-knot nematode infection and mi-mediated nematode resistance in tomato on plant fitness.

The Mi-1.2 resistance gene in tomato (Solanum lycopersicum) confers resistance against several species of root-knot nematodes (Meloidogyne spp.). This study examined the impact of M. javanica on the reproductive fitness of near-isogenic tomato cultivars with and without Mi-1.2 under field and greenhouse conditions. Surprisingly, neither nematode inoculation or host plant resistance impacted the yield of mature fruits in field microplots (inoculum=8,000 eggs/plant), or fruit or seed production in a follow-up greenhouse bioassay conducted with a higher inoculum level (20,000 eggs/plant). However, under heavy nematode pressure (200,000 eggs/plant), greenhouse-grown plants carrying Mi-1.2 had more than ten-fold greater fruit production than susceptible plants and nearly forty-fold greater estimated lifetime seed production, confirming prior reports of the benefits of Mi-1.2. In all cases Mi-mediated resistance significantly reduced nematode reproduction. These results indicated that tomato can utilize tolerance mechanisms to compensate for moderate levels of nematode infection, but that the Mi-1.2 resistance gene confers a dramatic fitness benefit under heavy nematode pressure. No significant cost of resistance was detected in the absence of nematode infection.