Different types of Bernstein operators in inference of Gaussian graphical model

Ağraz, Melih; Purutçuoğlu, Vi̇lda

doi:10.1080/23311835.2016.1154706

Cited by 1 publication

(1 citation statement)

References 23 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Because it is known that observations dependent on chemical master equations are described via the binomial distribution and any variation from these distributions can be caused by the nuisance effect, rather than the biological sources. On the other hand in order to control whether other Bernstein type operators such as Bleiman Butzer Hahn and Meyer‐König and the Zeller operator (Meyer–König & Zeller, ) show the same advantages we compare their precision with glasso and we find that Bernstein polynomials have the highest precision (Ağraz & Purutçuoğlu, ).…”

Section: Introductionmentioning

confidence: 99%

Bernstein approximations in glasso‐based estimation of biological networks

2017

Self Cite

View full text Add to dashboard Cite

The Gaussian graphical model (GGM) is one of the common dynamic modelling approaches in the construction of gene networks. In inference of this modelling the interaction between genes can be detected mainly via graphical lasso (glasso) or coordinate descent‐based approaches. Although these methods are successful in moderate networks, their performances in accuracy decrease when the system becomes sparser. We here implement a particular type of polynomial transformations, called the Bernstein polynomials, of the network data in advance of their inference to raise the accuracy. From comparative Monte Carlo studies and real data analyses we show that these polynomials are successful in terms of the precision, specificity and F‐measure when the scale‐free networks are modelled via GGM and estimated by glasso, and accordingly they can be used as a preprocessing step in inference of these networks. The Canadian Journal of Statistics 45: 62–76; 2017 © 2017 Statistical Society of Canada

show abstract

Section: Introductionmentioning

confidence: 99%