id	author	title	date	pages	extension	mime	words	sentences	flesch	summary	cache	txt
cord-024290-8z6us7v4	Allen, Edward E.	Time Series Adjustment Enhancement of Hierarchical Modeling of Arabidopsis Thaliana Gene Interactions	2020-02-01		.txt	text/plain	3236	209	53	Network models of gene interactions, using time course gene transcript abundance data, are computationally created using a genetic algorithm designed to incorporate hierarchical Bayesian methods with time series adjustments. Second, the addition of time series adjustment to improve the independence of the model's residuals gives these techniques stronger statistical foundations. In complicated modeling situations (e.g., like ours where we need to obtain closed form likelihoods of DAGs within a hierarchical structure in order to produce posterior probabilities of edges), it is common to derive results as if there were non-correlated residuals, as we have done in previous work. The use of the time series adjusted next state Norris-Patton likelihood, along with a tailor-made genetic algorithm and Bayesian model averaging, allows for the rigorous estimation of posterior probabilities for all gene pair interactions. Using the transcript abundance data for 26 Arabidopsis thaliana genes stimulated by ACC, gene interaction models for a next state with and without time series adjustment were computationally created, shown in Fig. 3 .	./cache/cord-024290-8z6us7v4.txt	./txt/cord-024290-8z6us7v4.txt