MLP-Based Regression Prediction Model For Compound Bioactivity

Qin, Yongfei; Li, Chao; Shi, Xiaoyan; Wang, Weigang

doi:10.3389/fbioe.2022.946329

Cited by 16 publications

(5 citation statements)

References 28 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…MLP-Regressor is a supervised learning technique that effectively processes information through nonlinear regression by optimizing the squared error. This robust algorithm has already shown its usefulness for various applications [49,50]. It is used to verify the repetitive and accurate nature of our results in forecasting the spatial evolution of frequency component energy.…”

Section: Mlp-regressor Modelmentioning

confidence: 69%

A Multi-Approach Analysis for Monitoring Wave Energy Driven by Coastal Extremes

Matar,

Abcha,

Abroug

et al. 2024

Water

View full text Add to dashboard Cite

This research investigates the behavior and frequency evolution of extreme waves in coastal areas through a combination of physical modeling, spectral analysis, and artificial intelligence (AI) techniques. Laboratory experiments were conducted in a wave flume, deploying various wave spectra, including JONSWAP (γ = 7), JONSWAP (γ = 3.3), and Pierson–Moskowitz, using the dispersive focusing technique, covering a broad range of wave amplitudes. Wave characteristics were monitored using fifty-one gauges at distances between 4 m and 14 m from the wave generator, employing power spectral density (PSD) analysis to investigate wave energy subtleties. A spectral approach of discrete wavelets identified frequency components. The energy of the dominant frequency components, d5 and d4, representing the peak frequency (fp = 0.75 Hz) and its first harmonic (2fp = 1.5 Hz), respectively, exhibited a significant decrease in energy, while others increased, revealing potential correlations with zones of higher energy dissipation. This study underscores the repeatable and precise nature of results, demonstrating the Multilayer Perceptron (MLP) machine learning algorithm’s accuracy in predicting the energy of frequency components. The finding emphasizes the importance of a multi-approach analysis for effectively monitoring energy in extreme coastal waves.

show abstract

Section: Mlp-regressor Modelmentioning

confidence: 69%

A Multi-Approach Analysis for Monitoring Wave Energy Driven by Coastal Extremes

Matar,

Abcha,

Abroug

et al. 2024

Water

View full text Add to dashboard Cite

show abstract

“…Propagation of output signals from hidden layers and output prediction based on the recognized pattern in the output layer are determined by activation functions used in these layers. 25 In QSAR, MLPs were used in the prediction of toxicity, 26 bioactivity for antibreast cancer drug development, 27 blood–brain barrier permeability for discovery of CNS (central nervous system) therapeutic drugs, 28 and so forth. Rings, cyclic structures, and other molecular patterns are encompassed in the SMILES representation.…”

Section: Discussionmentioning

confidence: 99%

Identifying High-Quality Leads among Screened Anticancerous Compounds Using SMILES Representations

Peter,

Kalakoti,

Sundar

2024

ACS Omega

View full text Add to dashboard Cite

Cancer is a lethal disease that affects numerous people worldwide. Chemotherapy stands as one of the most effective treatment regimens to combat cancer. Nevertheless, anticancer drugs face a high failure rate due to safety and efficacy issues. Drug failure could be subdued by instigating drug leads with reduced toxicity and enhanced efficacy. Computer-aided drug discovery endorses drug leads in manoeuvring protein and ligand structures or representations. Simplified molecular input line entry system (SMILES) is a linear notation representing the threedimensional structure of a molecule using symbols and alphanumeric characters. SMILES representation hoards rings and scaffold structures in its depiction. Mining ring and scaffold patterns from molecular SMILES would assist in ascertaining biological properties based on molecular patterns. Moreover, the emergence of artificial intelligence (AI) technologies would accelerate identification of efficient anticancer drug leads. AI algorithms proclaimed for their pattern recognition ability could be employed for identifying molecular patterns from SMILES representation, thereby enabling property prediction. Consequently, we developed a multilayer perceptron (MLP) model for the prediction of anticancer activity using SMILES of NCI-60 cancer growth inhibition data. Furthermore, the top 8 frequent scaffolds were identified on preliminary analysis of cancer growth inhibition data and ChEMBL drugs. The developed MLP model classified anticancer and nonanticancer compounds with a classification accuracy of 0.92. Also, benchmarking of the developed model with machine learning algorithms exhibited better performance of the MLP model.

show abstract

“…A Multilayer Perceptron model (MLP) [ 34 ] is a standard fully connected neural network model. It consists of input and output layers and at least one hidden layer.…”

Section: Methodsmentioning

confidence: 99%

Machine learning methods for adult OSAHS risk prediction

Ge,

Wu,

et al. 2024

BMC Health Serv Res

View full text Add to dashboard Cite

Background Obstructive sleep apnea hypopnea syndrome (OSAHS) is a common disease that can cause multiple organ damage in the whole body. Our aim was to use machine learning (ML) to build an independent polysomnography (PSG) model to analyze risk factors and predict OSAHS. Materials and methods Clinical data of 2064 snoring patients who underwent physical examination in the Health Management Center of the First Affiliated Hospital of Shanxi Medical University from July 2018 to July 2023 were retrospectively collected, involving 24 characteristic variables. Then they were randomly divided into training group and verification group according to the ratio of 7:3. By analyzing the importance of these features, it was concluded that LDL-C, Cr, common carotid artery plaque, A1c and BMI made major contributions to OSAHS. Moreover, five kinds of machine learning algorithm models such as logistic regression, support vector machine, Boosting, Random Forest and MLP were further established, and cross validation was used to adjust the model hyperparameters to determine the final prediction model. We compared the accuracy, Precision, Recall rate, F1-score and AUC indexes of the model, and finally obtained that MLP was the optimal model with an accuracy of 85.80%, Precision of 0.89, Recall of 0.75, F1-score of 0.82, and AUC of 0.938. Conclusion We established the risk prediction model of OSAHS using ML method, and proved that the MLP model performed best among the five ML models. This predictive model helps to identify patients with OSAHS and provide early, personalized diagnosis and treatment options.

show abstract

MLP-Based Regression Prediction Model For Compound Bioactivity

Cited by 16 publications

References 28 publications

A Multi-Approach Analysis for Monitoring Wave Energy Driven by Coastal Extremes

A Multi-Approach Analysis for Monitoring Wave Energy Driven by Coastal Extremes

Identifying High-Quality Leads among Screened Anticancerous Compounds Using SMILES Representations

Machine learning methods for adult OSAHS risk prediction

Contact Info

Product

Resources

About