An official website of the United States government
Identifying local structure in molecular simulations is of utmost importance. The most common existing approach to identify local structure is to calculate some geometrical quantity referred to as an order parameter. In simple cases order parameters are physically intuitive and trivial to develop ( e.g. , ion-pair distance), however in most cases, order parameter development becomes a much more difficult endeavor ( e.g. , crystal structure identification). Using ideas from computer vision, we adapt a specific type of neural network called a PointNet to identify local structural environments in molecular simulations. A primary challenge in applying machine learning techniques to simulation is selecting the appropriate input features. This challenge is system-specific and requires significant human input and intuition. In contrast, our approach is a generic framework that requires no system-specific feature engineering and operates on the raw output of the simulations, i.e. , atomic positions. We demonstrate the method on crystal structure identification in Lennard-Jones (four different phases), water (eight different phases), and mesophase (six different phases) systems. The method achieves as high as 99.5% accuracy in crystal structure identification. The method is applicable to heterogeneous nucleation and it can even predict the crystal phases of atoms near external interfaces. We demonstrate the versatility of our approach by using our method to identify surface hydrophobicity based solely upon positions and orientations of surrounding water molecules. Our results suggest the approach will be broadly applicable to many types of local structure in simulations.
more »
« less
Structurally Constrained Evolutionary Algorithm for the Discovery and Design of Metastable Phases
Metastable materials are abundant in nature and technology, showcasing remarkable properties that inspire innovative materials design. However, traditional crystal structure prediction methods, which rely solely on energetic factors to determine a structure’s fitness, are not suitable for predicting the vast number of potentially synthesizable phases that represent a local minimum corresponding to a state in thermodynamic equilibrium. Here, we present a new approach for the prediction of metastable phases with specific structural features and interface this method with the XTALOPT evolutionary algorithm. Our method relies on structural features that include the local crystalline order (e.g, the coordination number or chemical environment), and symmetry (e.g, Bravais lattice and space group) to filter the breeding pool of an evolutionary crystal structure search. The effectiveness of this approach is benchmarked on three known metastable systems: XeN8, with a two-dimensional polymeric nitrogen sublattice, brookite TiO2, and a high pressure BaH4 phase, which was recently characterized. Additionally, a newly predicted metastable melaminate salt, P1̅ WC3N6, was found to possess an energy that is lower than that of two phases proposed in a recent computational study. The method presented here could help in identifying the structures of compounds that have already been synthesized, and in developing new synthesis targets with desired properties.
more »
« less
- PAR ID:
- 10520711
- Publisher / Repository:
- Journal of Chemical Theory and Computation
- Date Published:
- Journal Name:
- Journal of Chemical Theory and Computation
- Volume:
- 19
- Issue:
- 21
- ISSN:
- 1549-9618
- Page Range / eLocation ID:
- 7960 to 7971
- Format(s):
- Medium: X
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
Recent experimental and theoretical work has demonstrated significant potential to tune the properties of silicon and germanium by adjusting the mesostructure, nanostructure, and/or crystalline structure of these group 14 elements. Despite the promise to achieve enhanced functionality with these already technologically important elements, a significant challenge lies in the identification of effective synthetic approaches that can access metastable silicon and germanium-based extended solids with a particular crystal structure or specific nano/meso-structured features. In this context, the class of intermetallic compounds known as Zintl phases has provided a platform for discovery of novel silicon and germanium-based materials. This review highlights some of the ways in which silicon and germanium-based Zintl phases have been utilized as precursors in innovative approaches to synthesize new crystalline modifications, nanoparticles, nanosheets, and mesostructured and nanoporous extended solids with properties that can be very different from the ground states of the elements.more » « less
-
Metastable phases were formed in Mn1−xCoxNiGe (x=0.05 and 0.08) by annealing at 800 °C followed by rapid cooling, i.e., quenching, at ambient pressure (P=0) and under a pressure of P=3.5 GPa, and their phase transitions and associated magnetocaloric properties were investigated. The crystal cell volumes of the metastable phases decreased, and their structural transitions significantly shifted to lower temperatures relative to those of the slow-cooled compounds, with a greater reduction observed in the samples where the rapid cooling occurred under high pressures. The magnetic and structural transitions coupled to form a magnetostructural transition in the metastable phases, resulting in large magnetic entropy changes up to −79.6 J kg−1 K−1 (x=0.08) for a 7-T field change. The experimental results demonstrate thermal quenching and high-pressure annealing as alternative methods to create magnetostructural transitions, without modifying the compositions of the materials.more » « less
-
Abstract MotivationQuality assessment (QA) of predicted protein tertiary structure models plays an important role in ranking and using them. With the recent development of deep learning end-to-end protein structure prediction techniques for generating highly confident tertiary structures for most proteins, it is important to explore corresponding QA strategies to evaluate and select the structural models predicted by them since these models have better quality and different properties than the models predicted by traditional tertiary structure prediction methods. ResultsWe develop EnQA, a novel graph-based 3D-equivariant neural network method that is equivariant to rotation and translation of 3D objects to estimate the accuracy of protein structural models by leveraging the structural features acquired from the state-of-the-art tertiary structure prediction method—AlphaFold2. We train and test the method on both traditional model datasets (e.g. the datasets of the Critical Assessment of Techniques for Protein Structure Prediction) and a new dataset of high-quality structural models predicted only by AlphaFold2 for the proteins whose experimental structures were released recently. Our approach achieves state-of-the-art performance on protein structural models predicted by both traditional protein structure prediction methods and the latest end-to-end deep learning method—AlphaFold2. It performs even better than the model QA scores provided by AlphaFold2 itself. The results illustrate that the 3D-equivariant graph neural network is a promising approach to the evaluation of protein structural models. Integrating AlphaFold2 features with other complementary sequence and structural features is important for improving protein model QA. Availability and implementationThe source code is available at https://github.com/BioinfoMachineLearning/EnQA. Supplementary informationSupplementary data are available at Bioinformatics online.more » « less
-
2D layered metal-organic frameworks (MOFs) are a new class of multifunctional materials that can provide electrical conductivity on top of the conventional structural characteristics of MOFs, offering potential applications in electronics and optics. Here, for the first time, we employ Machine Learning (ML) techniques to predict the thermodynamic stability and electronic properties of layered electrically conductive (EC) MOFs, bypassing expensive ab initio calculations for the design and discovery of new materials. Proper feature engineering is a very important factor in utilizing ML models for such purposes. Here, we show that a combination of elemental features, using generic statistical reduction methods and crystal structure information curated from the recently introduced EC-MOF database, leads to a reasonable prediction of the thermodynamic and electronic properties of EC MOFs. We utilize these features in training a diverse range of ML classifiers and regressors. Evaluating the performance of these different models, we show that an ensemble learning approach in the form of stacking ML models can lead to higher accuracy and more reliability on the predictive power of ML to be employed in future MOF research.more » « less
- Free Publicly Accessible Full Text
-
Accepted Manuscript1.0
- Journal Article:
- https://doi.org/10.1021/acs.jctc.3c00594
