AI Article Synopsis
- Molecular dynamics (MD) is widely used in structural biology to study the structure and function of macromolecules.
- Boltzmann generators offer a new method by using generative neural networks to simulate molecular systems more quickly than traditional MD, but they currently face theoretical and computational challenges.
- This study establishes a solid mathematical basis for Boltzmann generators, proving they can effectively replace MD in certain scenarios involving complex macromolecules like proteins, and introduces tools for navigating molecular energy landscapes with neural networks.
Article Abstract
Molecular dynamics (MD) is the primary computational method by which modern structural biology explores macromolecule structure and function. Boltzmann generators have been proposed as an alternative to MD, by replacing the integration of molecular systems over time with the training of generative neural networks. This neural network approach to MD enables convergence to thermodynamic equilibrium faster than traditional MD; however, critical gaps in the theory and computational feasibility of Boltzmann generators significantly reduce their usability. Here, we develop a mathematical foundation to overcome these barriers; we demonstrate that the Boltzmann generator approach is sufficiently rapid to replace traditional MD for complex macromolecules, such as proteins in specific applications, and we provide a comprehensive toolkit for the exploration of molecular energy landscapes with neural networks.
Download full-text PDF |
Source |
|---|---|
| http://www.ncbi.nlm.nih.gov/pmc/articles/PMC10720392 | PMC |
| http://dx.doi.org/10.1093/bib/bbad456 | DOI Listing |
Publication Analysis
Top Keywords
Similar Publications
Exploring Effects of Mental Stress with Data Augmentation and Classification Using fNIRS.
Sensors (Basel)
January 2025
Department of Electrical Engineering, American University of Sharjah, Sharjah 26666, United Arab Emirates.
Accurately identifying and discriminating between different brain states is a major emphasis of functional brain imaging research. Various machine learning techniques play an important role in this regard. However, when working with a small number of study participants, the lack of sufficient data and achieving meaningful classification results remain a challenge.
View Article and Find Full Text PDFRestoration of Genuine Sensation and Proprioception of Individual Fingers Following Transradial Amputation with Targeted Sensory Reinnervation as a Mechanoneural Interface.
J Clin Med
January 2025
Department of Trauma Surgery, Hannover Medical School, Carl-Neuberg-Str. 1, 30625 Hannover, Germany.
: Tactile gnosis derives from the interplay between the hand's tactile input and the memory systems of the brain. It is the prerequisite for complex hand functions. Impaired sensation leads to profound disability.
View Article and Find Full Text PDFEvaluation of Structure Prediction and Molecular Docking Tools for Therapeutic Peptides in Clinical Use and Trials Targeting Coronary Artery Disease.
Int J Mol Sci
January 2025
Department of Applied Biotechnology, Faculty of Chemistry, Warsaw University of Technology, 00-661 Warsaw, Poland.
This study evaluates the performance of various structure prediction tools and molecular docking platforms for therapeutic peptides targeting coronary artery disease (CAD). Structure prediction tools, including AlphaFold 3, I-TASSER 5.1, and PEP-FOLD 4, were employed to generate accurate peptide conformations.
View Article and Find Full Text PDFAutonomous learning of generative models with chemical reaction network ensembles.
J R Soc Interface
January 2025
Indian Institute of Technology Bombay, Mumbai, Maharashtra, India.
Can a micron-sized sack of interacting molecules autonomously learn an internal model of a complex and fluctuating environment? We draw insights from control theory, machine learning theory, chemical reaction network theory and statistical physics to develop a general architecture whereby a broad class of chemical systems can autonomously learn complex distributions. Our construction takes the form of a chemical implementation of machine learning's optimization workhorse: gradient descent on the relative entropy cost function, which we demonstrate can be viewed as a form of integral feedback control. We show how this method can be applied to optimize any detailed balanced chemical reaction network and that the construction is capable of using hidden units to learn complex distributions.
View Article and Find Full Text PDFLattice thermal conductivity in CrSBr: the effects of interlayer interaction, magnetic ordering and external strain.
J Phys Condens Matter
January 2025
South China Normal University, School of Physics, Guangdong Provincial Key Laboratory of Quantum Engineering and Quantum Materials, Guangdong Basic Research Center of Excellence for Structure and Fundamental Interactions of Matter, Guangzhou, 510631, CHINA.
With the continuous development of digital information and big data technologies, the ambient temperature and heat generation during the operation of magnetic storage devices play an increasingly crucial role in ensuring data security and device stability. In this study, we examined the lattice thermal conductivity of the van der Waals magnetic semiconductor CrSBr from bulk to monolayer structures using first-principles calculations and the phonon Boltzmann transport equation. Our results indicated that lattice thermal conductivity show anisotropy and CrSBr bilayer exhibits lower thermal conductivity at all temperatures.
View Article and Find Full Text PDFWant AI Summaries of new PubMed Abstracts delivered to your In-box?
Enter search terms and have AI summaries delivered each week - change queries or unsubscribe any time!