Structure Learning in Nested Effects Models - Quantitative Biology > Quantitative MethodsReport as inadecuate




Structure Learning in Nested Effects Models - Quantitative Biology > Quantitative Methods - Download this document for free, or read online. Document in PDF available to download.

Abstract: Nested Effects Models NEMs are a class of graphical models introduced toanalyze the results of gene perturbation screens. NEMs explore noisy subsetrelations between the high-dimensional outputs of phenotyping studies, e.g. theeffects showing in gene expression profiles or as morphological features of theperturbed cell.In this paper we expand the statistical basis of NEMs in four directions:First, we derive a new formula for the likelihood function of a NEM, whichgeneralizes previous results for binary data. Second, we prove modelidentifiability under mild assumptions. Third, we show that the new formulationof the likelihood allows to efficiently traverse model space. Fourth, weincorporate prior knowledge and an automated variable selection criterion todecrease the influence of noise in the data.



Author: Achim Tresch, Florian Markowetz

Source: https://arxiv.org/



DOWNLOAD PDF




Related documents