A framework for predicting odor threshold values of perfumes by scientific machine learning and transfer learning.

Knowledge of odor thresholds is very important for the perfume industry. Due to the difficulty associated with measuring odor thresholds, empirical models capable of estimating these values can be an invaluable contribution to the field. This work developed a framework based on scientific machine learning strategies. A transfer learning-based strategy was devised, where information from a graph convolutional network predicting semantic odor descriptors was used as input data for the feedforward neural network responsible for estimating odor thresholds for chemical substances based on their molecular structures. The predictive performance of this model was compared to a benchmark odor threshold prediction model based on molecular structures that did not utilize transfer learning. Furthermore, the prediction was compared to a correlation previously proposed in the literature and a dummy regressor. Results demonstrated that the transfer learning-based strategy displayed a better predictive performance, suggesting this technique can be useful for predicting odor thresholds.

Metaheuristic Framework for Material Screening and Operating Optimization of Adsorption-Based Heat Pumps.

The current methods applied to material screening for adsorption-based heat pumps are based on a fixed set of temperatures or their independent variation, providing a limited, insufficient, and unpractical evaluation of different adsorbents. This work proposes a novel strategy for the simultaneous optimization and material screening in the design of adsorption heat pumps by implementing a meta-heuristic approach, particle swarm optimization (PSO). The proposed framework can effectively evaluate variable and broad operation temperature intervals to search for viable zones of operation for multiple adsorbents at once. The criteria for selecting the adequate material were the maximum performance and the minimum heat supply cost, which were considered the objective functions of the PSO algorithm. First, the performance was assessed individually, followed by a single-objective approximation of the multi-objective problem. Next, a multi-objective approach was also adopted. With the results generated during the optimization, it was possible to find which adsorbents and temperature sets were the most suitable according to the main objective of the operation. The Fisher-Snedecor test was applied to expand the results obtained during PSO application and a feasible operating region built around the optima, enabling the arrangement of close-to-optima data into practical design and control tools. This approach allowed for a fast and intuitive evaluation of multiple design and operation variables.

Transient Coatings from Nanoparticles Achieving Broad-Spectrum and High Antimicrobial Performance.

Cationic and hydrophilic coatings based on casting and drying water dispersions of two different nanoparticles (NPs) onto glass are here described and evaluated for antimicrobial activity. Discoid cationic bilayer fragments (BF) surrounded by carboxy-methylcellulose (CMC) and poly (diallyl dimethyl ammonium) chloride (PDDA) NPs and spherical gramicidin D (Gr) NPs dispersed in water solution were cast onto glass coverslips and dried, forming a coating quantitatively evaluated against Pseudomonas aeruginosa, Staphylococcus aureus and Candida albicans. From plating and colony forming units (CFU) counting, all strains interacting for 1 h with the coatings lost viability from 105 to 106, to zero CFU, at two sets of Gr and PDDA doses: 4.6 and 25 µg, respectively, or, 0.94 and 5 µg, respectively. Combinations produced broad spectrum, antimicrobial coatings; PDDA electrostatically attached to the microbes damaging cell walls, allowing Gr NPs interaction with the cell membrane. This concerted action promoted optimal activity at low Gr and PDDA doses. Further washing and drying of the deposited dried coatings showed that they were washed out so that antimicrobial activity was no longer present on the glass surface. Significant applications in biomedical materials can be foreseen for these transient coatings.

Generating Flavor Molecules Using Scientific Machine Learning.

Flavor is an essential component in the development of numerous products in the market. The increasing consumption of processed and fast food and healthy packaged food has upraised the investment in new flavoring agents and consequently in molecules with flavoring properties. In this context, this work brings up a scientific machine learning (SciML) approach to address this product engineering need. SciML in computational chemistry has opened paths in the compound's property prediction without requiring synthesis. This work proposes a novel framework of deep generative models within this context to design new flavor molecules. Through the analysis and study of the molecules obtained from the generative model training, it was possible to conclude that even though the generative model designs the molecules through random sampling of actions, it can find molecules that are already used in the food industry, not necessarily as a flavoring agent, or in other industrial sectors. Hence, this corroborates the potential of the proposed methodology for the prospecting of molecules to be applied in the flavor industry.

A Reinforcement Learning Framework to Discover Natural Flavor Molecules.

Flavor is the focal point in the flavor industry, which follows social tendencies and behaviors. The research and development of new flavoring agents and molecules are essential in this field. However, the development of natural flavors plays a critical role in modern society. Considering this, the present work proposes a novel framework based on scientific machine learning to undertake an emerging problem in flavor engineering and industry. It proposes a combining system composed of generative and reinforcement learning models. Therefore, this work brings an innovative methodology to design new flavor molecules. The molecules were evaluated regarding synthetic accessibility, the number of atoms, and the likeness to a natural or pseudo-natural product. This work brings as contributions the implementation of a web scraper code to sample a flavors database and the integration of two scientific machine learning techniques in a complex system as a framework. The implementation of the complex system instead of the generative model by itself obtained 10% more molecules within the optimal results. The designed molecules obtained as an output of the reinforcement learning model's generation were assessed regarding their existence or not in the market and whether they are already used in the flavor industry or not. Thus, we corroborated the potentiality of the framework presented for the search of molecules to be used in the development of flavor-based products.

Novel Framework for Simulated Moving Bed Reactor Optimization Based on Deep Neural Network Models and Metaheuristic Optimizers: An Approach with Optimality Guarantee.

Model-based optimization of simulated moving bed reactors (SMBRs) requires efficient solvers and significant computational power. Over the past years, surrogate models have been considered for such computationally demanding optimization problems. In this sense, artificial neural networks-ANNs-have found applications for modeling the simulated moving bed (SMB) unit but not yet been reported for the reactive SMB (SMBR). Despite ANNs' high accuracy, it is essential to assess its capacity to represent the optimization landscape well. However, a consistent method for optimality assessment using surrogate models is still an open issue in the literature. As such, two main contributions can be highlighted: the SMBR optimization based on deep recurrent neural networks (DRNNs) and the characterization of the feasible operation region. This is done by recycling the data points from a metaheuristic technique-optimality assessment. The results demonstrate that the DRNN-based optimization can address such complex optimization while meeting optimality.

Bovine serum albumin and myoglobin separation by size exclusion SMB.

The separation of the proteins Bovine Serum Albumin (BSA) and Myoglobin (Mb) was achieved by Size-Exclusion Simulated Moving Bed (SE-SMB) and performed experimentally in the FlexSMB® unit, an SMB unit designed and built in the Laboratory of Separation and Reaction Engineering. Before accomplishing the separation experiments in the mentioned unit, separation regions were computed by simulation based on a phenomenological mathematical model to determine appropriate operating conditions. The developed model was validated in advance, against fixed-bed dynamic adsorption experimental results, for pure component and binary mixtures. Then the SMB experiments were carried out, and purities of the Mb on the extract and BSA on the raffinate streams were 98% and 96%, respectively. The achieved recoveries were 80% of Mb on the extract and 94% of BSA on the raffinate. Lastly, productivities of 6.4 gprotein⋅lads-1⋅day-1 for the extract and 28.8 gprotein⋅lads-1⋅day-1 for the raffinate were obtained.

Separation of tartronic and glyceric acids by simulated moving bed chromatography.

The SMB unit developed by the Laboratory of Separation and Reaction Engineering (FlexSMB-LSRE®) was used to perform tartronic acid (TTA) and glyceric acid (GCA) separation and to validate the mathematical model in order to determine the optimum operating parameters of an industrial unit. The purity of the raffinate and extract streams in the experiments performed were 80% and 100%, respectively. The TTA and GCA productivities were 79 and 115 kg per liter of adsorbent per day, respectively and only 0.50 cubic meters of desorbent were required per kilogram of products. Under the optimum operating conditions, which were determined through an extensive simulation study based on the mathematical model developed to predict the performance of a real SMB unit, it was possible to achieve a productivity of 86 kg of TTA and 176 kg of GCA per cubic meter of adsorbent per day (considering the typical commercial purity value of 97% for both compounds) with an eluent consumption of 0.30 cubic meters per kilogram of products.

Magnetically triggered release of amoxicillin from xanthan/Fe3O4/albumin patches.

This work was motivated by the need of stimuli responsive drug carriers, which can be activated by low cost non-invasive stimuli such as external magnetic field (EMF). Thus, novel antimicrobial materials based on xanthan gum (XG), magnetic nanoparticles (MNP), bovine serum albumin (BSA) and amoxicillin (Amox) were designed in order to promote the release of Amox under magnetic stimuli. Firstly, surfaces with different functionalities were prepared by sequential deposition of thin layers on Si wafers and characterized by means of ellipsometry and atomic force microscopy. Amox adsorbed preferentially onto XG or BSA films. In solution, favorable interactions between Amox and BSA were evidenced by substantial changes in the BSA secondary structure, as revealed by circular dichroism. Patches of XG and XG/MNP/BSA were immersed in 2 g L-1 Amox, yielding 10 ±â€¯3 and 17 ±â€¯4 µg/cm3 Amox loading, respectively. The inclusion of 0.2 wt% Fe3O4 in the patches and their exposure to EMF enabled in vitro release of Amox, at pH 5.5 and 0.02 mol L-1 NaCl, following the quasi-Fickian behavior. Amox diffused from XG/MNP/BSA patches in agar medium containing Staphylococcus aureus and Escherichia coli, inhibiting their growth. The inhibition of E. coli growth was particularly efficient under EMF.

Cationic Biomimetic Particles of Polystyrene/Cationic Bilayer/Gramicidin for Optimal Bactericidal Activity.

Nanostructured particles of polystyrene sulfate (PSS) covered by a cationic lipid bilayer of dioctadecyldimethylammonium bromide (DODAB) incorporated gramicidin D (Gr) yielding optimal and broadened bactericidal activity against both Escherichia coli and Staphylococcus aureus. The adsorption of DODAB/Gr bilayer onto PSS nanoparticles (NPs) increased the zeta-average diameter by 8-10 nm, changed the zeta-potential of the NPs from negative to positive, and yielded a narrow size distributions for the PSS/DODAB/Gr NPs, which displayed broad and maximal microbicidal activity at very small concentrations of the antimicrobials, namely, 0.057 and 0.0057 mM DODAB and Gr, respectively. The results emphasized the advantages of highly-organized, nanostructured, and cationic particles to achieve hybrid combinations of antimicrobials with broad spectrum activity at considerably reduced DODAB and Gr concentrations.

Involvement of the p62/NRF2 signal transduction pathway on erythrophagocytosis.

Erythrophagocytosis, the phagocytic removal of damaged red blood cells (RBC), and subsequent phagolysosome biogenesis are important processes in iron/heme metabolism and homeostasis. Phagolysosome biogenesis implies the interaction of nascent phagosomes with endocytic compartments and also autophagy effectors. Here, we report that besides recruitment of microtubule-associated protein-1-light chain 3 (LC3), additional autophagy machinery such as sequestosome 1 (p62) is also acquired by single-membrane phagosomes at very early stages of the phagocytic process and that its acquisition is very important to the outcome of the process. In bone marrow-derived macrophages (BMDM) silenced for p62, RBC degradation is inhibited. P62, is also required for nuclear translocation and activation of the transcription factor Nuclear factor E2-related Factor 2 (NRF2) during erythrophagocytosis. Deletion of the Nrf2 allele reduces p62 expression and compromises RBC degradation. In conclusion, we reveal that erythrophagocytosis relies on an interplay between p62 and NRF2, potentially acting as protective mechanism to maintain reactive oxygen species at basal levels and preserve macrophage homeostasis.

Dynamics of a True Moving Bed separation process: Linear model identification and advanced process control.

The control of Simulated Moving Bed (SMB) units is challenging due to their complex dynamic behaviour and the difficulty of measuring their main properties. Furthermore, for the SMB units, the transfer function identification when the unit is operating at its optimal point is not easy to be done through the usual way. This work presents the development of a novel strategy to identify transfer functions of TMB/SMB and its application on classical linear model predictive controllers (MPC). However, for the process in study, due its unique dynamics, only the identification of the linear model is not enough to solve its control problem. Therefore, it is proposed a modification in the MPC prediction, that consists in a strategy based on a switching system where the most adequate transfer function is employed in the controller to overcome the problems related with the process dynamic behaviour. The results show that the used methodology enables the easy identification of transfer functions at the process optimal operating point and that the MPC can control the process in both the servo and regulator problem cases. It is also showed that the transfer function identified can be applied in the control of a SMB unit with four columns, under its optimal conditions.

Sustainable hydroxypropyl methylcellulose/xyloglucan/gentamicin films with antimicrobial properties.

Hydroxypropyl methylcellulose (HPMC) and xyloglucan (XG) crosslinked with citric acid over a range of HPMC/XG weight ratios formed sustainable blend films characterized by Fourier transform infrared spectroscopy, thermogravimetric analysis, scanning electron microscopy, tensile tests, circular dichroism and determination of inhibitory activity against Staphylococcus aureus and Escherichia coli. Both in solution and in the crosslinked films, HPMC chains lost the original ordered conformation upon interacting with XG, giving rise to an entropic gain. The highest values of tensile strength (25MPa) and Young's modulus (689MPa) occurred for the 50:50 HPMC/XG blend films. In vitro loading of gentamicin sulfate (GS) in the films amounted to 0.18±0.05 -0.37±0.05g of GS per g polymer. At pH 7.4 and 37°C, the GS release kinetics from the films fitted with the Korsmeyer-Peppas model revealed a non-Fickian release mechanism with diffusional coefficient n∼0.7. The cross-linked films of HPMC, XG and their blends loaded with GS showed outstanding antibacterial activity against Staphylococcus aureus and Escherichia coli, disclosing their potential for novel biomedical applications.

Biomimetic Cationic Nanoparticles Based on Silica: Optimizing Bilayer Deposition from Lipid Films.

The optimization of bilayer coverage on particles is important for a variety of biomedical applications, such as drug, vaccine, and genetic material delivery. This work aims at optimizing the deposition of cationic bilayers on silica over a range of experimental conditions for the intervening medium and two different assemblies for the cationic lipid, namely, lipid films or pre-formed lipid bilayer fragments. The lipid adsorption on silica in situ over a range of added lipid concentrations was determined from elemental analysis of carbon, hydrogen, and nitrogen and related to the colloidal stability, sizing, zeta potential, and polydispersity of the silica/lipid nanoparticles. Superior bilayer deposition took place from lipid films, whereas adsorption from pre-formed bilayer fragments yielded limiting adsorption below the levels expected for bilayer adsorption.

Evaluation of prognostic factors and survival rates in malignant feline mammary gland neoplasms.

OBJECTIVES: The aim of the study was to investigate prognostic factors in feline mammary gland neoplasms, correlating them with overall survival (OS). METHODS: Fifty-six primary malignant mammary gland neoplasms and 16 metastatic lymph nodes from 37 female cats were analyzed. Clinical staging, histologic type and grade, and immunohistochemistry for Ki-67, progesterone and estrogen receptor, human epidermal growth factor receptor type 2 (HER-2), cyclooxygenase-2 (COX-2) and vascular endothelial growth factor (VEGF) were evaluated. Follow-up was performed in order to correlate prognostic factors with OS. RESULTS: Lymph node metastasis was found in 35% of cases. Clinical stage III, tubulopapillary carcinomas and histologic grade II cases prevailed in the study. Most neoplasms were positive for hormonal receptors, negative for HER-2 overexpression and presented VEGF overexpression. Immunoreactivity for Ki-67 (P = 0.046) and COX-2 (P = 0.007) was higher in metastases than in primary tumors. COX-2 (P = 0.089), HER-2 (P = 0.012) and histologic grade (P = 0.080) were correlated with OS. CONCLUSIONS AND RELEVANCE: The data suggest that inhibition of ovarian hormones and COX-2 may represent a therapeutic option for malignant feline mammary gland neoplasms. When evaluating disease progression, COX-2 scores and Ki-67 index should be analyzed in primary tumors and metastases. Histologic grade, HER-2 status and COX-2 scores were found to have a direct influence on OS. Prognostic factors allow for a better understanding of disease outcome in a condition that is characterized by a poor prognosis. The present work highlights the need for further studies on endocrine therapy and COX-2 inhibitors, which could influence OS.

Nrf2 as a master regulator of tissue damage control and disease tolerance to infection.

Damage control refers to those actions made towards minimizing damage or loss. Depending on the context, these can range from emergency procedures dealing with the sinking of a ship or to a surgery dealing with severe trauma or even to an imaginary company in Marvel comics, which repairs damaged property arising from conflicts between super heroes and villains. In the context of host microbe interactions, tissue damage control refers to an adaptive response that limits the extent of tissue damage associated with infection. Tissue damage control can limit the severity of infectious diseases without interfering with pathogen burden, conferring disease tolerance to infection. This contrasts with immune-driven resistance mechanisms, which although essential to protect the host from infection, can impose tissue damage to host parenchyma tissues. This damaging effect is countered by stress responses that confer tissue damage control and disease tolerance to infection. Here we discuss how the stress response regulated by the transcription factor nuclear factor-erythroid 2-related factor 2 (Nrf2) acts in such a manner.

Gas-phase simulated moving bed: Propane/propylene separation on 13X zeolite.

In the last years several studies were carried out in order to separate gas mixtures by SMB technology; however, this technology has never been implemented on an industrial scale. In the present work, a gas phase SMB bench unit was built and tested for the separation of propane and propylene mixtures, using 13X zeolite extrudates as adsorbent and isobutane as desorbent. Three experiments were performed to separate propane/propylene by gas phase SMB in the bench scale unit with a 4-2-2 configuration, i.e., open loop circuit by suppressing section IV (desorbent regeneration followed by a recycle). Consequently, all the experiments were conducted using an external supply of pure isobutane as desorbent. Parameters such as switching time, extract and raffinate stream flow rates were changed to improve the efficiency of the process. Experimental results have shown that it is feasible to separate propylene from propane by gas phase SMB at a bench scale and that this process is a potential candidate to replace the conventional technologies for the propane/propylene separation. The performance parameters obtained are very promising for future development of this technology, since propylene was obtained in the extract stream with a purity of 99.93%, a recovery of 99.51%, and a productivity of [Formula: see text] . Propane was obtained in the raffinate stream with a purity of 98.10%, a recovery of 99.73% and a productivity of [Formula: see text] . The success of the above mentioned bench scale tests is a big step for the future implementation of this technology in a larger scale.

Nanostructures for peroxidases.

Peroxidases are enzymes catalyzing redox reactions that cleave peroxides. Their active redox centers have heme, cysteine thiols, selenium, manganese, and other chemical moieties. Peroxidases and their mimetic systems have several technological and biomedical applications such as environment protection, energy production, bioremediation, sensors and immunoassays design, and drug delivery devices. The combination of peroxidases or systems with peroxidase-like activity with nanostructures such as nanoparticles, nanotubes, thin films, liposomes, micelles, nanoflowers, nanorods and others is often an efficient strategy to improve catalytic activity, targeting, and reusability.