Molecular Dynamics (MD)-Derived Features for Canonical and Noncanonical Amino Acids

Journal of Chemical Information and Modeling, Journal Year: 2025, Volume and Issue: unknown

Published: Feb. 2, 2025

Machine learning (ML) models have become increasingly popular for predicting and designing structures properties of peptides proteins. These ML typically use proteins containing only canonical amino acids as the training data. Consequently, these struggle to make accurate predictions new that are absent in data set (e.g., noncanonical acids). One approach improve accuracy is collect more with desired acids. However, this strategy suboptimal may not be easily attainable, additional time required retrain models. Alternatively, extendibility can improved if acid features used representative generalizable unseen Herein, we develop using molecular dynamics (MD) simulation results. Specifically, a given acid, perform MD its dipeptide create based on backbone (ϕ, ψ) distributions electrostatic potentials. We demonstrate enable our accurately predict structural ensembles cyclic present original set. For example, build pentapeptide structures, library 15 test same 15-amino-acid or an extended 50-amino-acid library. When such Morgan fingerprints MACCS keys represent acids, achieve R2 = 0.963 pentapeptides models' performances decrease significantly 0.430 0.508, respectively, when tasked 50 On other hand, model outperforms those keys, 0.700. Overall, instead having data, peptide sequences originally at mere cost performing simulations

Language: Английский

---

Machine Learning Approaches in Metabolic Pathway Predictions and Drug-Target Interactions: Advancing Drug Discovery

Published: Jan. 1, 2025

Language: Английский

---

Predicting amyloid proteins using attention-based long short-term memory

PeerJ Computer Science, Journal Year: 2025, Volume and Issue: 11, P. e2660 - e2660

Published: Feb. 7, 2025

Alzheimer's disease (AD) is one of the genetically inherited neurodegenerative disorders that mostly occur when people get old. It can be recognized by severe memory impairment in late stage, affecting cognitive function and general daily living. Reliable evidence confirms enhanced symptoms AD are linked to accumulation amyloid proteins. The dense population proteins forms insoluble fibrillar structures, causing significant pathological impacts various tissues. Understanding protein's mechanisms identifying them at an early stage plays essential role treating as well prevalent amyloid-related diseases. Recently, although several machine learning methods proposed for protein identification have shown promising results, most not yet fully exploited sequence information In this study, we develop a computational model silico using bidirectional long short-term combination with attention mechanism. testing phase, our findings showed developed method outperformed those state-of-the-art area under receiver operating characteristic curve 0.9126.

Language: Английский

---

Molecular docking technology drives multidimensional applications of microbial natural products

Journal of Molecular Structure, Journal Year: 2025, Volume and Issue: unknown, P. 142044 - 142044

Published: March 1, 2025

Language: Английский

---

Introduction to Cheminformatics for Predictive Modeling

Challenges and advances in computational chemistry and physics, Journal Year: 2025, Volume and Issue: unknown, P. 13 - 42

Published: Jan. 1, 2025

Language: Английский

---

QMGBP-DL: a deep learning and machine learning approach for quantum molecular graph band-gap prediction

Molecular Diversity, Journal Year: 2025, Volume and Issue: unknown

Published: April 19, 2025

Language: Английский

---

From Patterns to Pills: How Informatics Is Shaping Medicinal Chemistry

Pharmaceutics, Journal Year: 2025, Volume and Issue: 17(5), P. 612 - 612

Published: May 5, 2025

In today’s information-driven era, machine learning is revolutionizing medicinal chemistry, offering a paradigm shift from traditional, intuition-based, and often bias-prone methods to the prediction of chemical properties without prior knowledge basic principles governing drug function. This perspective highlights growing importance informatics in shaping field particularly through concept “informacophore”. The informacophore refers minimal structure, combined with computed molecular descriptors, fingerprints, machine-learned representations its that are essential for molecule exhibit biological activity. Similar skeleton key unlocking multiple locks, points features trigger responses. By identifying optimizing informacophores in-depth analysis ultra-large datasets potential lead compounds automating standard parts development process, there will be significant reduction biased intuitive decisions, which may systemic errors parallel acceleration discovery processes.

Language: Английский

---

From SMILES to Enhanced Molecular Property Prediction: A Unified Multimodal Framework with Predicted 3D Conformers and Contrastive Learning Techniques

Journal of Chemical Information and Modeling, Journal Year: 2024, Volume and Issue: unknown

Published: Dec. 6, 2024

We present a novel molecular property prediction framework that requires only the SMILES format as input but is designed to be multimodal by incorporating predicted 3D conformer representations. Our model captures comprehensive features leveraging both sequential character structure of and three-dimensional spatial conformers. The employs contrastive learning techniques, utilizing InfoNCE loss align embeddings, along with task-specific functions, such ConR for regression SupCon classification. To address data imbalance, we incorporate feature distribution smoothing (FDS), common challenge in drug discovery. evaluated through multiple case studies, including SARS-CoV-2 docking score prediction, using MoleculeNet sets, kinase inhibitor JAK-1, JAK-2, MAPK-14 custom sets curated from PubChem. results consistently outperformed state-of-the-art methods, FDS significantly improving tasks enhancing classification performance. These findings highlight flexibility robustness our model, demonstrating its effectiveness across diverse tasks, promising applications discovery analysis.

Language: Английский

---

Applications of Transformers in Computational Chemistry: Recent Progress and Prospects

The Journal of Physical Chemistry Letters, Journal Year: 2024, Volume and Issue: unknown, P. 421 - 434

Published: Dec. 31, 2024

The powerful data processing and pattern recognition capabilities of machine learning (ML) technology have provided technical support for the innovation in computational chemistry. Compared with traditional ML deep (DL) techniques, transformers possess fine-grained feature-capturing abilities, which are able to efficiently accurately model dependencies long-sequence data, simulate complex diverse chemical spaces, explore logic behind data. In this Perspective, we provide an overview application transformer models We first introduce working principle analyze transformer-based architectures Next, practical applications a number specific scenarios such as property prediction structure generation. Finally, based on these research results, outlook field future.

Language: Английский

---

Identifying Nephrotoxicity of Small Molecules Using Machine Learning

TENCON 2021 - 2021 IEEE Region 10 Conference (TENCON), Journal Year: 2024, Volume and Issue: unknown, P. 482 - 485

Published: Dec. 1, 2024

Language: Английский

---