Variable Selection and Joint Estimation of Mean and Covariance Models with an Application to eQTL Data

Lee, JungJun; Kim, Sunghwan; Jhong, Jae-Hwan; Koo, Ja−Yong

doi:10.1155/2018/4626307

Cited by 3 publications

(1 citation statement)

References 32 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In recent years, joint variable selection and covariance estimation in high-dimensional settings has generated a lot of interest (Banterle et al, 2018;Consonni, Rocca & Peluso, 2017;Deshpande, Ročková & George, 2019). Examples include the multi-omics data such as eQTL and metabolomics analysis, where gene expression and metabolomics data are simultaneously available and analyzed (see Lee et al, 2018;Nica & Dermitzakis, 2013;Rockman & Kruglyak, 2006).…”

Section: Introductionmentioning

confidence: 99%

Joint variable selection and network modeling for detecting eQTLs

Cao

Ding

Mersha

2020

Statistical Applications in Genetics and Molecular Biology

View full text Add to dashboard Cite

In this study, we conduct a comparison of three most recent statistical methods for joint variable selection and covariance estimation with application of detecting expression quantitative trait loci (eQTL) and gene network estimation, and introduce a new hierarchical Bayesian method to be included in the comparison. Unlike the traditional univariate regression approach in eQTL, all four methods correlate phenotypes and genotypes by multivariate regression models that incorporate the dependence information among phenotypes, and use Bayesian multiplicity adjustment to avoid multiple testing burdens raised by traditional multiple testing correction methods. We presented the performance of three methods (MSSL -Multivariate Spike and Slab Lasso, SSUR -Sparse Seemingly Unrelated Bayesian Regression, and OBFBF -Objective Bayes Fractional Bayes Factor), along with the proposed, JDAG (Joint estimation via a Gaussian Directed Acyclic Graph model) method through simulation experiments, and publicly available HapMap real data, taking asthma as an example. Compared with existing methods, JDAG identified networks with higher sensitivity and specificity under row-wise sparse settings. JDAG requires less execution in small-to-moderate dimensions, but is not currently applicable to high dimensional data. The eQTL analysis in asthma data showed a number of known gene regulations such as STARD3, IKZF3 and PGAP3, all reported in asthma studies. The code of the proposed method is freely available at GitHub (https://github.com/xuan-cao/Joint-estimation-for-eQTL).

show abstract

Section: Introductionmentioning

confidence: 99%

Joint variable selection and network modeling for detecting eQTLs

Cao

Ding

Mersha

2020

Statistical Applications in Genetics and Molecular Biology

View full text Add to dashboard Cite

show abstract

Variable Selection in Joint Mean and Covariance Models

Kou

Pan

2020

Recent Developments in Multivariate and Random Matrix Analysis

View full text Add to dashboard Cite

Modelling speciation: Problems and implications

Bard

2023

ISB

View full text Add to dashboard Cite

Darwin’s and Wallace’s 1859 explanation that novel speciation resulted from natural variants that had been subjected to selection was refined over the next 150 years as genetic inheritance and the importance of mutation-induced change were discovered, the quantitative theory of evolutionary population genetics was produced, the speed of genetic change in small populations became apparent and the ramifications of the DNA revolution became clear. This paper first discusses the modern view of speciation in its historical context. It then uses systems-biology approaches to consider the many complex processes that underpin the production of a new species; these extend in scale from genes to populations with the processes of variation, selection and speciation being affected by factors that range from mutation to climate change. Here, events at a particular scale level (e.g. protein network activity) are activated by the output of the level immediately below (i.e. gene expression) and generate a new output that activates the layer above (e.g. embryological development), with this change often being modulated by feedback from higher and lower levels. The analysis shows that activity at each level in the evolution of a new species is marked by stochastic activity, with mutation of course being the key step for variation. The paper examines events at each scale level and particularly considers how the pathway by which mutation leads to phenotypic variants and the wide range of factors that drive selection can be investigated computationally. It concludes that, such is the complexity of speciation, most steps in the process are currently difficult to model and that predictions about future speciation will, apart from a few special cases, be hard to make. The corollary is that opportunities for novel variants to form are maximised.

show abstract

Variable Selection and Joint Estimation of Mean and Covariance Models with an Application to eQTL Data

Cited by 3 publications

References 32 publications

Joint variable selection and network modeling for detecting eQTLs

Joint variable selection and network modeling for detecting eQTLs

Variable Selection in Joint Mean and Covariance Models

Modelling speciation: Problems and implications

Contact Info

Product

Resources

About