pAtbP-EnC: Identifying Anti-Tubercular Peptides Using Multi-Feature Representation and Genetic Algorithm-Based Deep Ensemble Model

IEEE Access, Год журнала: 2023, Номер 11, С. 137099 - 137114

Опубликована: Янв. 1, 2023

Mycobacterium tuberculosis, a highly perilous pathogen in humans, serves as the causative agent of tuberculosis (TB), affecting nearly 33% global population. With increasing prevalence multidrug-resistant TB, there is needs for novel and efficacious alternative therapies. Peptide therapies have emerged favorable due to its remarkable specificity targeting effected cells without effecting healthy cells. However, experimental identification anti-tubercular peptides (AtbPs) labor-intensive costly. Therefore, accurate prediction AtbPs has become challenging large number peptide samples. In this paper, we propose an ensemble learning model enhance outcomes by addressing limitations individual models. We formulate training samples utilizing four distinct representation methods: AAindex, Composition/Transition/Distribution, Dipeptide Deviation from Expected Mean, Enhanced Grouped Amino Acid Composition numerically encode The feature vectors extracted these methods are fused develop compact vector. evaluate rates using three different classification models, employing both heterogeneous vectors. Furthermore, capabilities proposed predicted labels classifiers implementing deep via genetic algorithm. Through evaluation on datasets independent datasets, our learner achieves impressive accuracies 97.80%, 95.13%, 93.91%, 94.17%, RD training, MD independent, respectively. Our findings demonstrate that pAtbP-EnC outperforms existing predictors reporting approximately 11% higher accuracy. conclude predictor will be considerable tool field pharmaceutical design research academia. used source code publicly available at https://github.com/Intelligent-models/pAtbP-EnC2023.

Язык: Английский

---

Finite element, analytical, artificial neural network models for carbon fibre reinforced polymer confined concrete filled steel columns with elliptical cross sections

Frontiers in Materials, Год журнала: 2023, Номер 9

Опубликована: Янв. 18, 2023

In the present era of architecture, different cross-sectional shapes structural concrete elements have been utilized. However, this change in shape has a significant effect on load-carrying capacity. To restore this, use column confinements with elliptical sections gained attention. This paper aim to investigate confined reinforced Carbon Fiber Reinforced Polymer (CFRP) and steel tube axial study is achieved using following tools Finite Element (FE) Abaqus Artificial Neural Networks (ANN) modeling. The involved 500-mm-high three sets aspect ratios: 1.0, 1.5, 2.0. each ratio, layers CFRP were used, i.e., .167, .334, .501-mm. Analytical results showed that increase ratio from 1 2, there decrease ultimate load about 23.2% average. addition, combined confining pressure increases dilation angle as number increases. failure mode for large local buckling at its mid-height along minor axis. result good correlation between FE experimental stress strains, mean squared error 2.27 .001, respectively. Moreover, ANN analytical models delightful R 2 .97 .88 strain models, section tubes can be adopted new architectural type construction; however, more than ratios, wrapping jackets highly recommended.

Язык: Английский

---

Predictive modelling of compressive strength of fly ash and ground granulated blast furnace slag based geopolymer concrete using machine learning techniques

Case Studies in Construction Materials, Год журнала: 2024, Номер 20, С. e03130 - e03130

Опубликована: Апрель 4, 2024

Ordinary Portland cement (OPC) is proving to be hazardous the environment. To replace OPC, geopolymers (GPs) are introduced. However, fully OPC by GPs extensive laboratory tests required assess long-term and short-term properties of in different scenarios. Given shortage time for performing such testing, artificial intelligence (AI) used analyze GPs. In this study, AI techniques as neuro network (ANN), adaptive neuro-fuzzy inference system (ANFIS), gene expression programming (GEP) obtain predictive models estimating compressive strength fly ash ground granulated blast furnace slag-based GP concrete. Different statistical parameters evaluate performance models. Similarly, sensitivity parametric analysis also conducted on input parameters. Additionally, multiple linear regression was performed whole database. After comparing all results, it concluded that GEP best technique predict GP-based

Язык: Английский

---

Computational prediction of workability and mechanical properties of bentonite plastic concrete using multi-expression programming

Scientific Reports, Год журнала: 2024, Номер 14(1)

Опубликована: Март 13, 2024

Abstract Bentonite plastic concrete (BPC) demonstrated promising potential for remedial cut-off wall construction to mitigate dam seepage, as it fulfills essential criteria strength, stiffness, and permeability. High workability consistency are attributes BPC because is poured into trenches using a tremie pipe, emphasizing the importance of accurately predicting slump BPC. In addition, prediction models offer valuable tools estimate various strength parameters, enabling adjustments mixing designs optimize project construction, leading cost time savings. Therefore, this study explores multi-expression programming (MEP) technique predict key characteristics BPC, such slump, compressive ( fc ), elastic modulus Ec ). present study, 158, 169, 111 data points were collected from experimental studies , Ec, respectively. The dataset was divided three sets: 70% training, 15% testing, another model validation. MEP exhibited excellent accuracy with correlation coefficient (R) 0.9999 0.9831 fc, 0.9300 Ec. Furthermore, comparative analysis between conventional linear non-linear regression revealed remarkable precision in predictions proposed models, surpassing traditional methods. SHapley Additive exPlanation indicated that water, cement, bentonite exert significant influence on water having greatest impact while curing cement exhibit higher modulus. summary, application machine learning algorithms offers capability deliver prompt precise early estimates properties, thus optimizing efficiency design processes.

Язык: Английский

---

Forecasting the strength characteristics of concrete incorporating waste foundry sand using advance machine algorithms including deep learning

Case Studies in Construction Materials, Год журнала: 2023, Номер 19, С. e02459 - e02459

Опубликована: Сен. 9, 2023

The incorporation of waste foundry sand (WFS) into concrete has been recognized as a sustainable approach to improve the strength properties (WFSC). However, machine learning (ML) techniques are still necessary forecast characteristics WFSC and evaluate dominant input features for suitable mix design. For this purpose, present work selected five ML-based based on gene expression programming (GEP), deep neural network (DNN), optimizable Gaussian process regressor (OGPR) predict mechanical WFSC. To build up predictive models, database containing 397 values compressive (CS) 169 flexural (FS) is collected from published literature. models' performance was evaluated via various statistical metrics additionally, external validation criteria were employed validate developed models. Furthermore, Shapley additive explanation (SHAP) carried out interpret model's prediction. DNN2 model exhibited superior performance, with R-values 0.996 (training), 0.999 (testing), 0.997 (validation) estimation. In contrast, GEP2 showed poor accuracy in estimating CS compared other 0.851, 0.901, 0.844 training, testing, sets, respectively. Similarly, estimation, provided indicating its robust performance. SHAP analysis revealed that age, water-cement ratio, coarse aggregate-to-cement ratio have prime influence determining strength, comparison models accurately estimated output high lower error might be utilized practical fields reduce labor cost by optimizing combinations. Finally, future studies, it recommended utilize ensemble hybrid algorithms, well post-hoc explanatory techniques, accurately.

Язык: Английский

---

Prediction of Mechanical Properties of Rubberized Concrete Incorporating Fly Ash and Nano Silica by Artificial Neural Network Technique

Axioms, Год журнала: 2023, Номер 12(1), С. 81 - 81

Опубликована: Янв. 12, 2023

The use of enormous amounts material is required for production. Due to the current emphasis on environment and sustainability materials, waste products by-products, including silica fume fly ash (FA), are incorporated into concrete as a substitute partially cement. Additionally, fine aggregate has indeed been largely replaced by materials like crumb rubber (CR), thus it reduces mechanical properties but improved some other concrete. To decrease detrimental effects CR, therefore enhanced with nanomaterials such nano (NS). essential designing constRuction structures. Concrete several variables can have its characteristics predicted an artificial neural network (ANN) technique. Using ANN approaches, this paper predict constructed FA partial cement, CR replacement aggregate, NS addition. technique, investigated comprise splitting tensile strength (Fs), compressive (Fc), modulus elasticity (Ec) flexural (Ff). model was used train test dataset obtained from experimental program. Fc, Fs, Ff Ec were added admixtures NS, curing age (P). modelling result indicated that high accuracy. proportional deviation mean (MoD) values calculated −0.28%, 0.14%, 0.87% 1.17%, respectively, which closed zero line. resulting model’s square error (MSE) coefficient determination (R2) 6.45 × 10−2 0.99496, respectively.

Язык: Английский

---

AI-powered COVID-19 forecasting: a comprehensive comparison of advanced deep learning methods

Osong Public Health and Research Perspectives, Год журнала: 2024, Номер 15(2), С. 115 - 136

Опубликована: Март 28, 2024

Objectives: The coronavirus disease 2019 (COVID-19) pandemic continues to pose significant challenges the public health sector, including that of United Arab Emirates (UAE). objective this study was assess efficiency and accuracy various deep-learning models in forecasting COVID-19 cases within UAE, thereby aiding nation’s authorities informed decision-making.Methods: This utilized a comprehensive dataset encompassing confirmed cases, demographic statistics, socioeconomic indicators. Several advanced deep learning models, long short-term memory (LSTM), bidirectional LSTM, convolutional neural network (CNN), CNN-LSTM, multilayer perceptron, recurrent (RNN) were trained evaluated. Bayesian optimization also implemented fine-tune these models.Results: evaluation framework revealed each model exhibited different levels predictive precision. Specifically, RNN outperformed other architectures even without optimization. Comprehensive perspective analytics conducted scrutinize dataset.Conclusion: transcends academic boundaries by offering critical insights enable UAE deploy targeted data-driven interventions. model, which identified as most reliable accurate for specific context, can significantly influence decisions. Moreover, broader implications research validate capability techniques handling complex datasets, thus transformative potential healthcare sectors.

Язык: Английский

---

Enhancing lung cancer detection through hybrid features and machine learning hyperparameters optimization techniques

Heliyon, Год журнала: 2024, Номер 10(4), С. e26192 - e26192

Опубликована: Фев. 1, 2024

Machine learning offers significant potential for lung cancer detection, enabling early diagnosis and potentially improving patient outcomes. Feature extraction remains a crucial challenge in this domain. Combining the most relevant features can further enhance detection accuracy. This study employed hybrid feature approach, which integrates both Gray-level co-occurrence matrix (GLCM) with Haralick autoencoder an autoencoder. These were subsequently fed into supervised machine methods. Support Vector (SVM) Radial Base Function (RBF) SVM Gaussian achieved perfect performance measures, while polynomial produced accuracy of 99.89% when utilizing GLCM autoencoder, Haralick, features. 99.56%, RBF 99.35% results demonstrate proposed approach developing improved diagnostic prognostic treatment planning decision-making systems.

Язык: Английский

---

Mathematical formulation for predicting moisture damage indices of asphalt mixtures treated with sustainable waste plastic modifiers using gene expression programming

Construction and Building Materials, Год журнала: 2024, Номер 426, С. 136146 - 136146

Опубликована: Апрель 13, 2024

Язык: Английский

---

Gene expression programming-based multivariate model for earth infrastructure: predicting ultimate bearing capacity of rock socketed shafts in layered soil-rock strata

Modeling Earth Systems and Environment, Год журнала: 2024, Номер 10(4), С. 5241 - 5256

Опубликована: Июль 3, 2024

Язык: Английский

---