Machine learning models for predicting interaction affinity energy between human serum proteins and hemodialysis membrane materials

Scientific Reports, Journal Year: 2025, Volume and Issue: 15(1)

Published: Jan. 28, 2025

Membrane incompatibility poses significant health risks, including severe complications and potential fatality. Surface modification of membranes has emerged as a pivotal technology in the membrane industry, aiming to improve hemocompatibility performance dialysis by mitigating undesired membrane-protein interactions, which can lead fouling subsequent protein adsorption. Affinity energy, defined strength interaction between human serum proteins, plays crucial role assessing interactions. These interactions may trigger adverse reactions, potentially harmful patients. Researchers often rely on trial-and-error approaches enhance reducing these This study focuses developing machine learning algorithms that accurately rapidly predict affinity energy novel chemical structures materials based molecular docking dataset. Various with distinct characteristics, chemistry, orientation are considered conjunction different proteins. A comparative analysis linear regression, K-nearest neighbors decision tree random forest XGBoost lasso support vector regression is conducted energy. The dataset, comprising 916 records for both training test segments, incorporates 12 parameters extracted from data points involves six Results indicate (R² = 0.8987, MSE 0.36, MAE 0.45) 0.83, 0.49, 0.49) exhibit comparable predictive However, outperforms testing Seven predicting analyzed compared, demonstrating superior accuracy. application holds promise researchers professionals hemodialysis. models, enabling early interventions hemodialysis membranes, could patient safety optimize care

Language: Английский

---

Deep learning accelerates reverse design of Magnetorheological elastomer

Composites Science and Technology, Journal Year: 2025, Volume and Issue: unknown, P. 111148 - 111148

Published: March 1, 2025

Language: Английский

---

Deep-Learning-Assisted Understanding of the Self-Assembly of Miktoarm Star Block Copolymers

ACS Nano, Journal Year: 2025, Volume and Issue: unknown

Published: March 12, 2025

The self-assemblies of topological complex block copolymers, especially the ABn type miktoarm star ones, are fascinating topics in soft matter field, which represent typical self-assembly behaviors analogous to those biological membranes. However, their diverse asymmetries and versatile spontaneous curvatures result rather phase separations that deviate significantly from common mechanisms. Thus, numerous trial-and-error experiments with tremendous parameter space intricate relationships needed study assemblies. Herein, we applied deep learning technology decipher copolymer PEO-s-PS2 an evaporation-induced system. A neural network model was trained practical experimental data encompassing two polymer properties three synthesis condition parameters as input variables, successfully predicted a three-dimensional (3D) synthesis-field diagram mined relationship between obtained structures. This demonstrated highly flexible structure modulation directions copolymer, revealing correlation parameters, conditions, output structures due significant influence variables on curvatures. work efficiency technique uncovering underlying rules systems, providing valuable insights into exploration science.

Language: Английский

---

Evolving Biomaterials Design from Trial and Error to Intelligent Innovation

Acta Biomaterialia, Journal Year: 2025, Volume and Issue: unknown

Published: March 1, 2025

Language: Английский

---

Machine Learning in Drug Development for Neurological Diseases: A Review of Blood Brain Barrier Permeability Prediction Models

Molecular Informatics, Journal Year: 2025, Volume and Issue: 44(3)

Published: March 1, 2025

The blood brain barrier (BBB) is an endothelial-derived structure which restricts the movement of certain molecules between general somatic circulatory system to central nervous (CNS). While BBB maintains homeostasis by regulating molecular environment induced cerebrovascular perfusion, it also presents significant challenges in developing therapeutics intended act on CNS targets. Many drug development practices rely partly extensive cell and animal models predict, extent, whether prospective therapeutic can cross BBB. In interest reduce costs improve prediction accuracy, many propose using advanced computational modeling permeability profiles leveraging empirical data. Given scale growth machine learning deep learning, we review most recent approaches predicting permeability.

Language: Английский

---