Predictive Computational Models Research Articles

Recent Advances in Artificial Intelligence to Improve Immunotherapy and the Use of Digital Twins to Identify Prognosis of Patients with Solid Tumors

To date, the public health system has been impacted by the increasing costs of many diagnostic and therapeutic pathways due to limited resources. At the same time, we are constantly seeking to improve these paths through approaches aimed at personalized medicine. To achieve the required levels of diagnostic and therapeutic precision, it is necessary to integrate data from different sources and simulation platforms. Today, artificial intelligence (AI), machine learning (ML), and predictive computer models are more efficient at guiding decisions regarding better therapies and medical procedures. The evolution of these multiparametric and multimodal systems has led to the creation of digital twins (DTs). The goal of our review is to summarize AI applications in discovering new immunotherapies and developing predictive models for more precise immunotherapeutic decision-making. The findings from this literature review highlight that DTs, particularly predictive mathematical models, will be pivotal in advancing healthcare outcomes. Over time, DTs will indeed bring the benefits of diagnostic precision and personalized treatment to a broader spectrum of patients.

Integrating forecasting methods to support finite element analysis and explore heat transfer complexities

Purpose This paper aims to reduce the cost of experiments required to test the efficiency of materials suitable for artificial tissue ablation by increasing efficiency and accurately forecasting heating properties. Design/methodology/approach A two-step numerical analysis is used to develop and simulate a bioheat model using improved finite element method and deep learning algorithms, systematically regulating temperature distributions within the hydrogel artificial tissue during radiofrequency ablation (RFA). The model connects supervised learning and finite element analysis data to optimize electrode configurations, ensuring precise heat application while protecting surrounding hydrogel integrity. Findings The model accurately predicts a range of thermal changes critical for optimizing RFA, thereby enhancing treatment precision and minimizing impact on surrounding hydrogel materials. This computational approach not only advances the understanding of thermal dynamics but also provides a robust framework for improving therapeutic outcomes. Originality/value A computational predictive bioheat model, incorporating deep learning to optimize electrode configurations and minimize collateral tissue damage, represents a pioneering approach in interventional research. This method offers efficient evaluation of thermal strategies with reduced computational overhead compared to traditional numerical methods.

Computational model for policy simulation and prediction of the regulatory impact of front-of-package food labels

Non-communicable diseases (NCDs), a significant cause of global mortality, necessitate targeted policy measures to reduce exposure to risk factors, particularly nutrition-related ones. Front-of-pack food labeling aims to shape consumer behavior and decrease exposure to nutritional risk factors. There are various forms of this policy, including traffic light (TL), nutrient profile model (NPM), and reference value (RV) labels. Mathematical models incorporating experimental and dose–response data support determining the most effective form. When selecting a modeling method, it is essential to consider the uncertainty involved in estimating risk factors. This study aims to create a computational model to predict the health and economic impacts of different policy options using an improved version of the Scarborough algorithm, incorporating a validation process to address potential uncertainties. The model shows that NPM was the most effective, reducing the estimated annual mortality rate from 514,247 cases to 447,394 cases (13% reduction). The following labels were TL (11% reduction) and RV (6% reduction). Economically, the RV label affects consumer spending the least, predicting an 8% increase in monthly spending per individual from the baseline. When assessed from both health and economic perspectives, the RV label has emerged as the most advantageous front-of-pack food labeling policy. The proposed model can aid policymakers in evaluating policy alternatives and thus selecting the most advantageous policies based on health and economic outcomes while ensuring the model's accuracy and applicability for the community.

An omics-driven computational model for angiogenic protein prediction: Advancing therapeutic strategies with Ens-deep-AGP

Angiogenic proteins (AGPs) play a critical role in both pathological and physiological activities, making them key therapeutic targets in diseases like cancer, heart disease, and stroke. Traditional methods for identifying AGPs are labor-intensive and time-consuming, creating a need for more efficient approaches. This study addresses this challenge by developing a novel computational model, Ens-Deep-AGP, designed to enhance AGP prediction. The model introduces innovative feature engineering techniques, including Position Specific Scoring Matrix-Decomposition-Discrete Cosine Transform (PSSM-DC-DCT) and Position Specific Scoring Matrix-Auto-Cross-Discrete Wavelet Transform (PSSM-ACC-DWT), which capture comprehensive protein sequence information. The ensemble feature set of these approaches are then fed into Multi-headed Ensemble Residual Convolutional Neural Network (MERCNN), a robust deep learning architecture. Ens-Deep-AGP achieved remarkable accuracy rates of 99.79 % on training dataset and 92.97 % on testing dataset, surpassing Convolutional Neural Networks (CNN), Gated Recurrent Units (GRU), and Bidirectional Long Short-Term Memory Networks (BiLSTM). The successful prediction of AGPs is crucial for accelerating drug development, discovering novel therapeutic targets and deepen our understanding of AGPs' complex roles in healthcare.

Galactic cosmic ray environment predictions for the NASA BioSentinel Mission, part 2:Post-mission validation

The BioSentinel CubeSat was deployed on the Artemis-I mission in November 2022 and has been continuously transmitting physical measurements of the space radiation environment since that time. Just before mission launch, we published computational model predictions of the galactic cosmic ray exposure expected inside BioSentinel for multiple locations and configurations. The predictions utilized models for the ambient galactic cosmic ray environment, radiation physics and transport, and BioSentinel geometry. Now that the nominal six-month BioSentinel mission has completed and some additional time has passed, those pre-launch predictions and additional model components can be validated. Dose-rate and linear energy transfer (LET) spectral measurements from the on-board dosimeter are presented along with a summary of the computational models used to calculate exposure quantities of interest. Sensitivity tests are performed to gauge the impact of various model choices on these quantities. Satellite data collected during the BioSentinel mission are used to provide some measure of independent validation for the galactic cosmic ray model used in the present calculations. It is shown that the combined models are in excellent agreement with the measured dose-rate. Model calculations agree well with measurement below ∼10 keV/µm and underpredict at higher LET. It is argued that the underprediction is likely due to detector response or low energy anomalous cosmic ray contributions able to reach the thinly shielded side of the on-board dosimeter.

Integrative Modeling of Signaling Network Dynamics Identifies Cell Type-Selective Therapeutic Strategies for FGFR4-Driven Cancers.

Oncogenic FGFR4 signaling represents a potential therapeutic target in various cancer types, including triple-negative breast cancer and hepatocellular carcinoma. However, resistance to FGFR4 single-agent therapy remains a major challenge, emphasizing the need for effective combinatorial treatments. Our study sought to develop a comprehensive computational model of FGFR4 signaling and to provide network-level insights into resistance mechanisms driven by signaling dynamics. An integrated approach, combining computational network modeling with experimental validation, uncovered potent AKT reactivation following FGFR4 targeting in triple-negative breast cancer cells. Analyzing the effects of cotargeting specific network nodes by systematically simulating the model predicted synergy of cotargeting FGFR4 and AKT or specific ErbB kinases, which was subsequently confirmed through experimental validation; however, cotargeting FGFR4 and PI3K was not synergistic. Protein expression data from hundreds of cancer cell lines was incorporated to adapt the model to diverse cellular contexts. This revealed that although AKT rebound was common, it was not a general phenomenon. For example, ERK reactivation occurred in certain cell types, including an FGFR4-driven hepatocellular carcinoma cell line, in which there is a synergistic effect of cotargeting FGFR4 and MEK but not AKT. In summary, this study offers key insights into drug-induced network remodeling and the role of protein expression heterogeneity in targeted therapy responses. These findings underscore the utility of computational network modeling for designing cell type-selective combination therapies and enhancing precision cancer treatment. Significance: Computational predictive modeling of signaling networks can decipher mechanisms of cancer cell resistance to targeted therapies and enable identification of more effective cancer type-specific combination treatment strategies.

Hybrid Class Balancing Approach for Chemical Compound Toxicity Prediction.

Computational methods are crucial for efficient and cost-effective drug toxicity prediction. Unfortunately, the data used for prediction is often imbalanced, resulting in biased models that favor the majority class. This paper proposes an approach to apply a hybrid class balancing technique and evaluate its performance on computational models for toxicity prediction in Tox21 datasets. The process begins by converting chemical compound data structures (SMILES strings) from various bioassay datasets into molecular descriptors that can be processed by algorithms. Subsequently, Undersampling and Oversampling techniques are applied in two different schemes on the training data. In the first scheme (Individual), only one balancing technique (Oversampling or Undersampling) is used. In the second scheme (Hybrid), the training data is divided according to a ratio (e.g., 90-10), applying a different balancing technique to each proportion. We considered eight resampling techniques (four Oversampling and four Undersampling), six molecular descriptors (based on MACCS, ECFP, and Mordred), and five classification models (KNN, MLP, RF, XGB and SVM) over 10 bioassay datasets to determine the configurations that yield the best performance. We defined three testing scenarios: without balancing techniques (baseline), Individual, and Hybrid. We found that using the ENN technique in the MACCS-MLP combination resulted in a 10.01% improvement in performance. The increase for ECFP6-2048 was 16.47% after incorporating a combination of the SMOTE (10%) and RUS (90%) techniques. Meanwhile, using the same combination of techniques, MORDRED-XGB showed the most significant increase in performance, achieving a 22.62% improvement. Integrating any of the class balancing schemes resulted in a minimum of 10.01% improvement in prediction performance compared to the best baseline configuration. In this study, Undersampling techniques were more appropriate due to the significant overlap among samples. By eliminating specific samples from the predominant class that are close to the minority class, this overlap is greatly reduced.

Modeling Cortical Versus Hippocampal Network Dysfunction in a Human Brain Assembloid Model of Epilepsy and Intellectual Disability.

Neurodevelopmental disorders often impair multiple cognitive domains. For instance, a genetic epilepsy syndrome might cause seizures due to cortical hyperexcitability and present with memory impairments arising from hippocampal dysfunction. This study examines how a single disorder differentially affects distinct brain regions by using human patient iPSC-derived cortical- and hippocampal-ganglionic eminence assembloids to model Developmental and Epileptic Encephalopathy 13 (DEE-13), a condition arising from gain-of-function mutations in the SCN8A gene. While cortical assembloids showed network hyperexcitability akin to epileptogenic tissue, hippocampal assembloids did not, and instead displayed network dysregulation patterns similar to in vivo hippocampal recordings from epilepsy patients. Predictive computational modeling, immunohistochemistry, and single-nucleus RNA sequencing revealed changes in excitatory and inhibitory neuron organization that were specific to hippocampal assembloids. These findings highlight the unique impacts of a single pathogenic variant across brain regions and establish hippocampal assembloids as a platform for studying neurodevelopmental disorders.

GLNNMDA: a multimodal prediction model for microbe-drug associations based on global and local features

Microbes have been demonstrated to be closely linked to diseases that pose a major threat to human health. Computing technologies can help researchers find potential microbe-drug associations more quickly and precisely. In this study, we introduced a novel computational prediction model called GLNNMDA based on global and local features of microbes and drugs to infer possible microbe-drug correlations. In GLNNMDA, we first constructed a heterogeneous network based on known microbe-drug relationships by integrating multiple similarity metrics of drugs and microbes. Subsequently, low-dimensional features will be extracted for nodes in the heterogeneous network by adopting the graph attention encoder. Next, based on combining these low-dimensional features with multiple properties of microbes and drugs to form a new comprehensive feature matrix, we would utilize the GLF module to extract the global and local features for microbes and drugs respectively, and then, we would further fuse these global and local features to come up with predictions of possible microbe-drug associations. Moreover, in order to evaluate the prediction performance of GLNNMDA, under the framework of fivefold cross-validation, intensive comparative experiments and case studies were done on different well-known public databases. The results showed that GLNNMDA obtained the highest AUC values as well as AUPR values of 0.9802 ± 0.0011, 0.9773 ± 0.0021 and 0.8586 ± 0.0004, 0.8008 ± 0.0031 in the two databases, MDAD and aBiofilm, respectively, compared to the state-of-the-art competing prediction methods. In addition, case studies of well-known microorganisms and drugs have demonstrated the effectiveness of GLNNMDA in inferring potential microbial drug associations, which implies that GLNNMDA may be a useful tool for microbe-drug association prediction in the future. The source code is available at: “https://github.com/KuangHaiYue/GLNNMDA.git”.

A general prediction model for compound-protein interactions based on deep learning.

The identification of compound-protein interactions (CPIs) is crucial for drug discovery and understanding mechanisms of action. Accurate CPI prediction can elucidate drug-target-disease interactions, aiding in the discovery of candidate compounds and effective synergistic drugs, particularly from traditional Chinese medicine (TCM). Existing in silico methods face challenges in prediction accuracy and generalization due to compound and target diversity and the lack of largescale interaction datasets and negative datasets for model learning. To address these issues, we developed a computational model for CPI prediction by integrating the constructed large-scale bioactivity benchmark dataset with a deep learning (DL) algorithm. To verify the accuracy of our CPI model, we applied it to predict the targets of compounds in TCM. An herb pair of Astragalus membranaceus and Hedyotis diffusaas was used as a model, and the active compounds in this herb pair were collected from various public databases and the literature. The complete targets of these active compounds were predicted by the CPI model, resulting in an expanded target dataset. This dataset was next used for the prediction of synergistic antitumor compound combinations. The predicted multi-compound combinations were subsequently examined through in vitro cellular experiments. Our CPI model demonstrated superior performance over other machine learning models, achieving an area under the Receiver Operating Characteristic curve (AUROC) of 0.98, an area under the precision-recall curve (AUPR) of 0.98, and an accuracy (ACC) of 93.31% on the test set. The model's generalization capability and applicability were further confirmed using external databases. Utilizing this model, we predicted the targets of compounds in the herb pair of Astragalus membranaceus and Hedyotis diffusaas, yielding an expanded target dataset. Then, we integrated this expanded target dataset to predict effective drug combinations using our drug synergy prediction model DeepMDS. Experimental assay on breast cancer cell line MDA-MB-231 proved the efficacy of the best predicted multi-compound combinations: Combination I (Epicatechin, Ursolic acid, Quercetin, Aesculetin and Astragaloside IV) exhibited a half-maximal inhibitory concentration (IC50) value of 19.41μM, and a combination index (CI) value of 0.682; and Combination II (Epicatechin, Ursolic acid, Quercetin, Vanillic acid and Astragaloside IV) displayed a IC50 value of 23.83μM and a CI value of 0.805. These results validated the ability of our model to make accurate predictions for novel CPI data outside the training dataset and evaluated the reliability of the predictions, showing good applicability potential in drug discovery and in the elucidation of the bioactive compounds in TCM. Our CPI prediction model can serve as a useful tool for accurately identifying potential CPI for a wide range of proteins, and is expected to facilitate drug research, repurposing and support the understanding of TCM.

Computational modeling and uncertainty prediction of hyperelastic constitutive responses of damaged brain tissue under different temperature and strain rates.

The effect of strain rate and temperature on the hyperelastic material stress-strain characteristics of the damaged porcine brain tissue is evaluated in this present work. The desired constitutive responses are obtained using the commercially available finite element (FE) tool ABAQUS, utilizing 8-noded brick elements. The model's accuracy has been verified by comparing the results from the previously published literature. Further, the stress-strain behavior of the brain tissue is evaluated by varying the damages at various strain rates and temperatures (13, 20, 27, and 37°C) under compression test. Additionally, the sensitivity analysis of the model is computed to check the effect of input parameters, that is, the temperature, strain rate, and damages on the material properties (shear modulus). The modeling and discussion sections enumerate the inclusive features and model capabilities.

Applying Machine Learning Techniques: Uncertainty Quantification in Nonlinear Dynamics Characters Predictions via Gated Recurrent Unit-Based Reduced-Order Models

The development of reduced-order models has been a pivotal advancement in the computational analysis of fluid dynamics, substantially simplifying the complexity and boosting the efficiency of simulations. The accuracy and practicality of these models largely depend on the reduction techniques applied and the inherent characteristics of the fluid dynamics systems they represent. In this paper, we introduce an innovative machine-learning framework for assessing model uncertainty in computationally intensive reduced-order models. By combining subspace construction methods with advanced Bayesian inference techniques, our approach effectively captures the posterior distribution of model parameters, thereby providing an accurate representation of uncertainty in aerodynamic performance predictions. We employ the NACA0012 airfoil as a case study to validate our method’s ability to enhance the efficiency of reduced-order models and precisely measure the uncertainty inherent in predictions made by recurrent neural networks. It is important to note that our approach is influenced by specific constraints and variables that significantly impact the mean and variability of the predicted final lift coefficient distribution. Our findings indicate that setting the goodness of fit (R2) threshold above 0.985 markedly improves the correlation between Computational Fluid Dynamics (CFD) outcomes and model predictions, increasing from 72.2% to 97.9% as the interval amplification factor adjusts from 1.5 to 3. However, this adjustment causes a considerable expansion of the confidence interval, from 0.0737 to 0.1282, an increase of over 70%. Despite these challenges, our machine learning-based methodology provides essential insights into the further development of reduced-order modeling and uncertainty quantification in fluid dynamics. This highlights the need for ongoing research into the model parameters, especially in applications concerning aircraft control systems, to meet design specifications and ensure system reliability.

Computational Hemodynamics-Based Growth Prediction for Small Abdominal Aortic Aneurysms: Laminar Simulations Versus Large Eddy Simulations.

Prior studies have shown that computational fluid dynamics (CFD) simulations help assess patient-specific hemodynamics in abdominal aortic aneurysms (AAAs); patient-specific hemodynamic stressors are frequently used to predict an AAA's growth. Previous studies have utilized both laminar and turbulent simulation models to simulate hemodynamics. However, the impact of different CFD simulation models on the predictive modeling of AAA growth remains unknown and is thus the knowledge gap that motivates this study. Specifically, CFD simulations were performed for 70 AAA models derived from 70 patients' computed tomography angiography (CTA) data with known growth status (i.e., fast-growing [> 5mm/yr] or slowly growing [< 5mm/yr]). We used laminar and large eddy simulation (LES) models to obtain hemodynamic parameters to predict AAAs' growth status. Predicting the growth status of AAAs was based on morphological, hemodynamic, and patient health parameters in conjunction with three classical machine learning (ML) classifiers, namely, support vector machine (SVM), K-nearest neighbor (KNN), and generalized linear model (GLM). Our preliminary results estimated aneurysmal flow stability and wall shear stress (WSS) were comparable in both laminar and LES flow simulations. Moreover, computed WSS and velocity-related hemodynamic variables obtained from the laminar and LES simulations showed comparable abilities in differentiating the growth status of AAAs. More importantly, the predictive modeling performance of the three ML classifiers mentioned above was similar, with less than a 2% difference observed (p-value>0.05). In closing, our findings suggest that two different flow simulations investigated did not significantly affect outcomes of computational hemodynamics and predictive modeling of AAAs' growth status, given the data investigated.

A computational predictive model for nanozyme diffusion dynamics: optimizing nanosystem performance

PurposeEnzymes play a pivotal role in orchestrating essential biochemical processes and influencing various cellular activities in tissue. This paper aims to provide the process of enzyme diffusion within the tissue matrix and enhance the nano system performance by means of the effectiveness of enzymatic functions. The diffusion phenomena are also documented, providing chemical insights into the complex processes governing enzyme movement.Design/methodology/approachA computational analysis is used to develop and simulate an optimal control model using numerical algorithms, systematically regulating enzyme concentrations within the tissue scaffold.FindingsThe accompanying videographic footages offer detailed insights into the dynamic complexity of the system, enriching the reader’s understanding. This comprehensive exploration not only contributes valuable knowledge to the field but also advances computational analysis in tissue engineering and biomimetic systems. The work is linked to biomolecular structures and dynamics, offering a detailed understanding of how these elements influence enzymatic functions, ultimately bridging the gap between theoretical insights and practical implications.Originality/valueA computational predictive model for nanozyme that describes the reaction diffusion dynamics process with enzyme catalysts is yet not available in existing literature.

Improvised grey wolf optimizer assisted artificial neural network (IGWO-ANN) predictive models to accurately predict the permeate flux of desalination plants

Effective planning, management, and control of industrial plants and processes have exploded in popularity to enhance global sustainability in recent decades. In this arena, computational predictive models have significantly contributed to plant performance optimization. In this regard, this research proposes an Improvised Grey Wolf Optimizer (IGWO) aided Artificial Neural Network (ANN) predictive model (IGWO-ANN Model-1 to 4) to predict the performance (permeate flux) of desalination plants accurately. For this, the proposed models investigated experimental inputs four: salt concentration & feed flow rate, condenser & evaporator inlet temperatures of the plant. Besides, mean squared error (MSE) and the regression coefficients (R2) have been used to assess the models' accuracy. The proposed IGWO-ANN Model-4 shows strong optimization abilities and provides better R2 = 99.3 % with minimum errors (0.004) compared to existing Response Surface Methodology (RSM) (R2 = 98.5 %, error = 0.100), ANN (R2 = 98.8 %, error = 0.060), GWO-ANN (R2 = 98.8 % error = 0.008), models. The proposed models are multitasking, multilayers, and multivariable, capable of accurately analyzing the desalination plant's performance, and suitable for other industrial applications. This study yielded a promising outcome and revealed the significant pathways for the researchers to analyze the desalination plant's performance to save time, money, and energy.

Pangenome reconstruction of Lactobacillaceae metabolism predicts species-specific metabolic traits.

Strains across the Lactobacillaceae family form the basis for a trillion-dollar industry. Our understanding of the genomic basis for their key traits is fragmented, however, including the metabolism that is foundational to their industrial uses. Pangenome analysis of publicly available Lactobacillaceae genomes allowed us to generate genome-scale metabolic network reconstructions for 26 species of industrial importance. Their manual curation led to more than 75,000 gene-protein-reaction associations that were deployed to generate 2,446 genome-scale metabolic models. Cross-referencing genomes and known metabolic traits allowed for manual metabolic network curation and validation of the metabolic models. As a result, we provide the first pangenomic basis for metabolism in the Lactobacillaceae family and a collection of predictive computational metabolic models that enable a variety of practical uses.IMPORTANCELactobacillaceae, a bacterial family foundational to a trillion-dollar industry, is increasingly relevant to biosustainability initiatives. Our study, leveraging approximately 2,400 genome sequences, provides a pangenomic analysis of Lactobacillaceae metabolism, creating over 2,400 curated and validated genome-scale models (GEMs). These GEMs successfully predict (i) unique, species-specific metabolic reactions; (ii) niche-enriched reactions that increase organism fitness; (iii) essential media components, offering insights into the global amino acid essentiality of Lactobacillaceae; and (iv) fermentation capabilities across the family, shedding light on the metabolic basis of Lactobacillaceae-based commercial products. This quantitative understanding of Lactobacillaceae metabolic properties and their genomic basis will have profound implications for the food industry and biosustainability, offering new insights and tools for strain selection and manipulation.

Finite elements for Matérn-type random fields: Uncertainty in computational mechanics and design optimization

This work highlights an approach for incorporating realistic uncertainties into scientific computing workflows based on finite elements, focusing on prevalent applications in computational mechanics and design optimization. We leverage Matérn-type Gaussian random fields (GRFs) generated using the SPDE method to model aleatoric uncertainties, including environmental influences, variating material properties, and geometric ambiguities. Our focus lies on delivering practical GRF realizations that accurately capture imperfections and variations and understanding how they impact the predictions of computational models as well as the shape and topology of optimized designs. We describe a numerical algorithm based on solving a generalized SPDE to sample GRFs on arbitrary meshed domains. The algorithm leverages established techniques and integrates seamlessly with the open-source finite element library MFEM and associated scientific computing workflows, like those found in industrial and national laboratory settings. Our solver scales efficiently for large-scale problems and supports various domain types, including surfaces and embedded manifolds. We showcase its versatility through biomechanics and topology optimization applications, emphasizing the potential to influence these domains. The flexibility and efficiency of SPDE-based GRF generation empowers us to run large-scale optimization problems on 2D and 3D domains, including finding optimized designs on embedded surfaces, and to generate design features and topologies beyond the reach of conventional techniques. Moreover, these capabilities allow us to model and quantify geometric uncertainties on reconstructed submanifolds, such as the interpolated surfaces of cerebral aneurysms provided by postprocessing CT scans. In addition to offering benefits in these specific domains, the proposed techniques transcend specific applications and generalize to arbitrary forward and backward problems in uncertainty quantification involving finite elements.

Beyond Static Planning: Computational Predictive Modeling to Avoid Coronary Artery Occlusion in TAVR

BackgroundCoronary artery occlusion (CO) during transcatheter aortic valve replacement (TAVR) is a devastating complication. The objective of this study was to assess the clinical impact of a computational predictive modeling algorithm for CO during TAVR planning. MethodsFrom January 2020 to December 2022, 116 patients (7.6%) who underwent TAVR evaluation were deemed to be at increased risk of CO on the basis of traditional criteria. Patients underwent prospective computational modeling (DASI Simulations) to assess their risk of CO during TAVR; procedural modifications and clinical results were reviewed retrospectively. ResultsOf the 116 patients at risk for CO by traditional methodology, 53 had native aortic valve stenosis (45.7%), 47 had undergone previous surgical AVR (40.5%), and 16 had undergone previous TAVR (13.8%). Transcatheter valve choice, size, or implantation depth was modeled for all patients. Computational modeling predicted an increased risk of CO in 39 of 116 (31.9%) patients. Within this subcohort, 29 patients proceeded with TAVR. Procedural modifications to augment the risk of CO included bioprosthetic or native aortic scallop intentional laceration to prevent iatrogenic coronary artery obstruction during TAVR (n = 10), chimney coronary stents (n = 8), and coronary access without stents (n = 3). There were no episodes of coronary artery compromise among patients after TAVR, either for those predicted to be at high risk of CO (with procedural modifications) or those predicted to be at low risk (standard TAVR). ConclusionsThe use of preoperative simulations for TAVR in patient-specific geometry through computational predictive modeling of CO is an effective enhancement to procedure planning.

Mathematical modeling of reverse osmosis system design and performance

ABSTRACT Reverse osmosis (RO) is one of the most effective technologies for water desalination. However, the RO system's performance is contingent on its design parameters and operating conditions. In this study, a computational model for RO desalination performance prediction was developed. The study compares the performance parameters of the RO system utilizing various types of membranes with the established model using a single pressure vessel and seven membrane elements within the pressure vessel. This study looked at two different feed water concentrations. The first is for saltwater with a concentration of 40,000 mg/L, while the second is for seawater with a concentration of 32,000 mg/L. WAVE software assessed the findings of the generated model to the recovery ratio, feed pressure, salt rejection, and permeate concentration of each membrane element inside the pressure vessel. The results of the study show good similarity between the simulation model results and the results obtained by WAVE software as they are illustrated in Figures 4–7.

ProkDBP: Toward more precise identification of prokaryotic DNA binding proteins.

Prokaryotic DNA binding proteins (DBPs) play pivotal roles in governing gene regulation, DNA replication, and various cellular functions. Accurate computational models for predicting prokaryotic DBPs hold immense promise in accelerating the discovery of novel proteins, fostering a deeper understanding of prokaryotic biology, and facilitating the development of therapeutics targeting for potential disease interventions. However, existing generic prediction models often exhibit lower accuracy in predicting prokaryotic DBPs. To address this gap, we introduce ProkDBP, a novel machine learning-driven computational model for prediction of prokaryotic DBPs. For prediction, a total of nine shallow learning algorithms and five deep learning models were utilized, with the shallow learning models demonstrating higher performance metrics compared to their deep learning counterparts. The light gradient boosting machine (LGBM), coupled with evolutionarily significant features selected via random forest variable importance measure (RF-VIM) yielded the highest five-fold cross-validation accuracy. The model achieved the highest auROC (0.9534) and auPRC (0.9575) among the 14 machine learning models evaluated. Additionally, ProkDBP demonstrated substantial performance with an independent dataset, exhibiting higher values of auROC (0.9332) and auPRC (0.9371). Notably, when benchmarked against several cutting-edge existing models, ProkDBP showcased superior predictive accuracy. Furthermore, to promote accessibility and usability, ProkDBP (https://iasri-sg.icar.gov.in/prokdbp/) is available as an online prediction tool, enabling free access to interested users. This tool stands as a significant contribution, enhancing the repertoire of resources for accurate and efficient prediction of prokaryotic DBPs.