The ReaxFF interatomic potential, used for organic materials, involves more than 600 adjustable parameters, the best-fit values of which must be determined for different materials. A new method of determining the set of best-fit parameters for specific molecules containing carbon, hydrogen, nitrogen and oxygen is presented, based on a parameter reduction technique followed by genetic algorithm (GA) minimization. This work has two novel features. The first is the use of a parameter reduction technique to determine which subset of parameters plays a significant role for the species of interest; this is necessary to reduce the optimization space to manageable levels. The second is the application of the GA technique to a complex potential (ReaxFF) with a very large number of adjustable parameters, which implies a large parameter space for optimization. In this work, GA has been used to optimize the parameter set to determine best-fit parameters that can reproduce molecular properties to within a given accuracy. As a test problem, the use of the algorithm has been demonstrated for nitromethane and its decomposition products.
Download full-text PDF |
Source |
|---|---|
| http://dx.doi.org/10.1007/s00894-011-1124-2 | DOI Listing |
Publication Analysis
Top Keywords
Similar Publications
Classification-based pathway analysis using GPNet with novel P-value computation.
Brief Bioinform
November 2024
Center for Artificial Intelligence Research, Wake Forest University School of Medicine, Winston-Salem, NC 27101, United States.
Pathway analysis plays a critical role in bioinformatics, enabling researchers to identify biological pathways associated with various conditions by analyzing gene expression data. However, the rise of large, multi-center datasets has highlighted limitations in traditional methods like Over-Representation Analysis (ORA) and Functional Class Scoring (FCS), which struggle with low signal-to-noise ratios (SNR) and large sample sizes. To tackle these challenges, we use a deep learning-based classification method, Gene PointNet, and a novel $P$-value computation approach leveraging the confusion matrix to address pathway analysis tasks.
View Article and Find Full Text PDFInferring the genetic relationships between unsupervised deep learning-derived imaging phenotypes and glioblastoma through multi-omics approaches.
Brief Bioinform
November 2024
Department of Neurosurgery, Peking Union Medical College Hospital, Chinese Academy of Medical Sciences and Peking Union Medical College, Beijing 100730, China.
This study aimed to investigate the genetic association between glioblastoma (GBM) and unsupervised deep learning-derived imaging phenotypes (UDIPs). We employed a combination of genome-wide association study (GWAS) data, single-nucleus RNA sequencing (snRNA-seq), and scPagwas (pathway-based polygenic regression framework) methods to explore the genetic links between UDIPs and GBM. Two-sample Mendelian randomization analyses were conducted to identify causal relationships between UDIPs and GBM.
View Article and Find Full Text PDFProtocol to boost the robustness and accuracy of spatial transcriptomics algorithms using ensemble techniques.
STAR Protoc
January 2025
Department of Statistics, University of Georgia, 310 Herty Drive, Athens, GA 30602, USA. Electronic address:
Spatial transcriptomics enhances our understanding of cellular organization by mapping gene expression data to precise tissue locations. Here, we present a protocol for using weighted ensemble method for spatial transcriptomics (WEST), which uses ensemble techniques to boost the robustness and accuracy of existing algorithms. We describe steps for preprocessing data, obtaining embeddings from individual algorithms, and ensemble integrating all embeddings as a similarity matrix.
View Article and Find Full Text PDFCovariate-adjusted construction of gene regulatory networks using a combination of generalized linear model and penalized maximum likelihood.
PLoS One
January 2025
Department of Computing and Mathematics, Manchester Metropolitan University, Manchester, United Kingdom.
Many machine learning techniques have been used to construct gene regulatory networks (GRNs) through precision matrix that considers conditional independence among genes, and finally produces sparse version of GRNs. This construction can be improved using the auxiliary information like gene expression profile of the related species or gene markers. To reach out this goal, we apply a generalized linear model (GLM) in first step and later a penalized maximum likelihood to construct the gene regulatory network using Glasso technique for the residuals of a multi-level multivariate GLM among the gene expressions of one species as a multi-levels response variable and the gene expression of related species as a multivariate covariates.
View Article and Find Full Text PDFThe impact of specialised gastroenterology services for pelvic radiation disease (PRD): Results from the prospective multi-centre EAGLE study.
PLoS One
January 2025
Marie Curie Research Centre, Division of Population Medicine, School of Medicine, Cardiff University, Cardiff, United Kingdom.
To undertake a mixed-methodology implementation study to improve the well-being of men with gastrointestinal late effects following radical radiotherapy for prostate cancer. All men completed a validated screening tool for late bowel effects (ALERT-B) and the Gastrointestinal Symptom Rating Score (GSRS); men with a positive score on ALERT-B were offered management following a peer reviewed algorithm for pelvic radiation disease (PRD). Health-related quality of life (HRQoL) at baseline, 6 and 12 months; and healthcare resource usage (HRU) and patient, support-giver, staff experience and acceptability of staff training (qualitative analysis) were assessed.
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!