A5037254330- Bayesian Methods and Mixture Models
- Statistical Methods and Bayesian Inference
- Protein Structure and Dynamics
- Enzyme Structure and Function
- Genetic and phenotypic traits in livestock
- Gaussian Processes and Bayesian Inference
- Genetic Mapping and Diversity in Plants and Animals
- Livestock Farming and Management
- Algorithms and Data Compression
- Machine Learning in Bioinformatics
- Genomics and Phylogenetic Studies
- Glycosylation and Glycoproteins Research
- Complex Network Analysis Techniques
- Meat and Animal Product Quality
- Human Mobility and Location-Based Analysis
- Effects of Environmental Stressors on Livestock
- Web Data Mining and Analysis
- Statistical Distribution Estimation and Applications
- Stochastic processes and statistical mechanics
- Statistical Methods and Inference
Commonwealth Scientific and Industrial Research Organisation
2017
Agriculture and Food
2017
Monash University
2014-2015
Indian Institute of Technology Madras
2010
Inbreeding has the potential to negatively impact animal performance. Strategies monitor and mitigate inbreeding depression require that it can be accurately estimated. Here, we used genomewide SNP data explore 3 alternative measures of genomic inbreeding: diagonal elements relationship matrix (FGRM), proportion homozygous (FHOM), genome covered by runs homozygosity (FROH). We 2,111 Brahman (BR) 2,550 Tropical Composite (TC) cattle with phenotypes recorded for 10 traits relevance tropical...
Numerical approaches to high-density single nucleotide polymorphism (SNP) data are often employed independently address individual questions. We linked independent in a bioinformatics pipeline for further insight. The driven by heterozygosity and Hardy-Weinberg equilibrium (HWE) analyses was applied characterize Bos taurus indicus ancestry. infer gene co-heterozygosity network that regulates bovine fertility, from on 18,363 cattle with genotypes 729,068 SNP. Hierarchical clustering separated...
The problem of superposition two corresponding vector sets by minimizing their sum-of-squares error under orthogonal transformation is a fundamental task in many areas science, notably structural molecular biology. This can be solved exactly using an algorithm whose time complexity grows linearly with the number correspondences. efficient solution has facilitated widespread use task, particularly studies involving macromolecular structures. article formally derives set sufficient statistics...
The modelling of data on a spherical surface requires the consideration directional probability distributions. To model asymmetrically distributed three-dimensional sphere, Kent distributions are often used. moment estimates parameters typically used in tasks involving However, these lack rigorous statistical treatment. focus paper is to introduce Bayesian estimation distribution which has not been carried out literature, partly because its complex mathematical form. We employ...
To compete with other telecom providers, it is important to understand the behavior of customers and predict their needs. In order realize this, required explore details based on mobile usage into social patterns (segments) target suitable segments for advertising. our approach, data in association browsing used form considered be an addition. From analysis rates respect a certain domain, operator can drill down sub domain level interests them specific customized services. This done by...
Proteins fold into complex three-dimensional shapes. Simplified representations of their shapes are central to rationalise, compare, classify, and interpret protein structures. Traditional methods abstract folding patterns rely on representing standard secondary structural elements (helices strands sheet) using line segments. This results in ignoring a significant proportion information. The motivation this research is derive mathematically rigorous biologically meaningful abstractions that...
The modelling of empirically observed data is commonly done using mixtures probability distributions. In order to model angular data, directional distributions such as the bivariate von Mises (BVM) typically used. critical task involved in mixture determine optimal number component We employ Bayesian information-theoretic principle minimum message length (MML) distinguish models by balancing trade-off between model's complexity and its goodness-of-fit data. consider problem resulting from...
Mixture modelling involves explaining some observed evidence using a combination of probability distributions. The crux the problem is inference an optimal number mixture components and their corresponding parameters. This paper discusses unsupervised learning models Bayesian Minimum Message Length (MML) criterion. To demonstrate effectiveness search parameters proposed approach, we select two key distributions, each handling fundamentally different types data: multivariate Gaussian...