Empowering Capacitive Devices: Harnessing Transfer Learning for Enhanced Data-Driven Optimization.

Developing data-driven models has found successful applications in engineering tasks, such as material design, process modeling, and process monitoring. In capacitive devices like deionization and supercapacitors, there exists potential for applying this data-driven machine learning (ML) model in optimizing its potential use in energy-efficient separations or energy generation. However, these models are faced with limited datasets, and even in large quantities, the datasets are incomplete, limiting their potential use for successful data-driven modeling. Here, the success of transfer learning in resolving the challenges with limited datasets was exploited. A two-step data-driven ML modeling framework named ImputeNet involving training with ML-imputed datasets and then with clean datasets was explored. Through data imputation and transfer learning, it is possible to develop a data-driven model with acceptable metrics mirroring experimental measurements. By using the model, optimization studies using the genetic algorithm were implemented to analyze the solution under the Pareto optimality. This early insight can be used in the initial stage of experimental measurements to rapidly identify experimental conditions worthy of further investigation. Moreover, we expect that the insights from these results will drive accurate predictive modeling in other fields including healthcare, genomic data analysis, and environmental monitoring with incomplete datasets.

A microfluidic approach to study variations in Chlamydomonas reinhardtii alkaline phosphatase activity in response to phosphate availability.

Algal growth depends strongly on phosphorus (P) as a key nutrient, underscoring the significance of monitoring P levels. Algal species display a sensitive response to fluctuations in P availability, notably through the expression of alkaline phosphatase (AP) when challenged with P-depletion. As such, alkaline phosphatase activity (APA) serves as a valuable metric for P availability, offering insights into how algae utilize and fix available P resources. However, current APA quantification methods lack single cell resolution, while also being time- and reagent consuming. Microfluidics offers a promising cost-effective solution to these limitations, providing a platform for precise single-cell analysis. In this study, a trap-based microfluidic device was integrated with a commercially available AP live stain to study the single cell APA response of a model algae strain, Chlamydomonas reinhardtii, when exposed to different exogenous P levels. A three-step culture-starve-spike process was used to induce APA in cells cultured under two different basal P levels (1 and 21 mM). When challenged with different spiked P levels (ranging from 0.1-41 mM), C. reinhardtii cells demonstrated a highly heterogeneous APA response. Two-way ANOVA confirmed that this response is influenced by both spiked and basal P levels. Utilizing an unsupervised machine learning approach (HDBSCAN), distinct subpopulations of C. reinhardtii cells were identified exhibiting varying levels of APA at the single-cell level. These subpopulations encompass significant groups of individual cells with either notably high or low APA, contributing to the overall behavior of the cohorts. Considerable intrapopulation differences in APA were observed across cohorts with similar average behavior. For instance, while some cohorts exhibited a concentrated distribution around the overall average APA, others displayed subpopulations dispersed across a wider range of APA levels. This underscores the potential bias introduced by analyzing a small number of cells in bulk, which may skew results by overrepresenting extreme behavioral subpopulations. The findings if this study highlight the need for analytical approaches that account for single cell heterogeneity in APA and demonstrate the utility of microfluidics as a well-suited means for such investigations. This study illuminates the complexities of APA regulation at the single cell level, providing crucial insights that advance our understanding of algal phosphorus metabolism and environmental responses.

Modular Framework for Simulation-Based Multi-objective Optimization of a Cryogenic Air Separation Unit.

A framework to obtain optimal operating conditions is proposed for a cryogenic air separation unit case study. The optimization problem is formulated considering three objective functions, 11 decision variables, and two constraint setups. Different optimization algorithms simultaneously evaluate the conflicting objective functions: the annualized cash flow, the efficiency at the compression stage, and capital expenditures. The framework follows a modular approach, in which the process simulator PRO/II and a Python environment are combined. The results permit us to assess the applicability of the tested algorithms and to determine optimal operational windows based on the resultant 3-D Pareto fronts.

Machine learning for guiding high-temperature PEM fuel cells with greater power density.

High-temperature polymer electrolyte membrane fuel cells (HT-PEMFCs) are enticing energy conversion technologies because they use low-cost hydrogen generated from methane and have simple water and heat management. However, proliferation of this technology requires improvement in power density. Here, we show that Machine Learning (ML) tools can help guide activities for improving HT-PEMFC power density because these tools quickly and efficiently explore large search spaces. The ML scheme relied on a 0-D, semi-empirical model of HT-PEMFC polarization behavior and a data analysis framework. Existing datasets underwent support vector regression analysis using a radial basis function kernel. In addition, the 0-D, semi-empirical HT-PEMFC model was substantiated by polarization data, and synthetic data generated from this model was subject to dimension reduction and density-based clustering. From these analyses, pathways were revealed to surpass 1 W cm-2 in HT-PEMFCs with oxygen as the oxidant and CO containing hydrogen.

Many-Objective Simulation-Based Optimization of an Air Separation Unit.

Air separation systems are crucial in the production of oxygen, which has gained particular relevance during the COVID-19 outbreak. Mechanical ventilation can compensate respiratory deficiencies along with the use of medical oxygen in vulnerable patients infected with this disease. In this contribution, a many-objective simulation-based optimization framework is proposed for determining eleven decision variables for the operation of an air separation unit. The framework combines the capabilities of the process simulator PRO/II with a Python environment. Three objective functions are optimized together towards the construction of a 3-D Pareto front. Results provide insightful information regarding the most adequate operating conditions of the unit, including the definition of an operational window rather than a single operational point.

Large-scale expansion of cytomegalovirus-specific cytotoxic T cells in suspension culture.

Clinical trials in recent years involving the adoptive transfer of antigen-specific cytotoxic T lymphocytes (CTL) have shown promise in restoring immunity against viral infection and reducing tumor burden in patients with solid and hematological malignancies. However, the large cell number required to achieve efficacy, 10(9) to 10(11), makes routine application of adoptive immunotherapy impractical. Investigation into new methods of CTL expansion may be useful in addressing this problem. Use of stirred suspension bioreactors are one such method that may allow large-scale T-cell expansion. Suspension cultures offer advantages over conventional static culture methods, including providing a homogeneous culture environment, and the potential for optimization and control of culture conditions. We generated cytomegalovirus (CMV)-specific CTL and investigated the potential of stirred bioreactor systems for expansion of large cell numbers. We found that CTL can be readily expanded ( > 200-fold) from cryopreserved stocks by nonspecific stimulation in the presence of allogeneic feeder cells and interleukin-2 (IL-2). Activated CTL inoculated into either suspension or static cultures could be subsequently expanded tenfold, and showed similar growth kinetics and metabolism independent of the culture vessel used. Furthermore, CTL remained specific for CMVpp65 peptide through the expansion phases, as demonstrated by pp65-tetramer staining ( > 95% tetramer(+)) and cytotoxicity assays. This study indicates that suspension reactor systems may be useful in large-scale expansion of antigen-specific CTL lines or clones, and may facilitate the advancement of routine adoptive immunotherapy.

Rapid, large-scale generation of highly pure cytomegalovirus-specific cytotoxic T cells for adoptive immunotherapy.

Adoptive transfer of donor-derived cytomegalovirus (CMV)-specific cytotoxic T cell (CTL) clones can restore immunity in allogeneic stem cell transplant recipients, providing protection against CMV disease. Current methods for selecting and expanding CMV-specific T cell clones are technically difficult, making adoptive T cell therapy impractical for routine clinical use. In this study, we describe a method for ex vivo generation and expansion of high-purity CMV-specific CTL using peptide-pulsed dendritic cells as antigen-presenting cells. Generation of CMV-specific CTL in numbers sufficient for clinical use in the time span of 4 weeks was accomplished in 6 of 8 CMV-seropositive donors. Examination of pp65 specificity by HLA/peptide tetramer staining demonstrated that a purity of greater than 95% peptide-specific cells could be obtained after two weekly stimulations and retained after further expansion for 3-4 weeks. Median expansion of total cell number was greater than 500-fold and expansion of peptide-specific CTL by tetramer staining was greater than 1.7 x 10(5)-fold. Four weeks after initiating CTL culture, we were able to generate greater than 10(9) total cells that specifically lysed target cells loaded with CMV peptide and cells infected with CMV. This simple and rapid method for generating high-purity CMV-specific CTL for adoptive immunotherapy is currently being examined for routine clinical use for allogeneic stem cell transplantation.