How to cluster gene expression dynamics in response to environmental signals

Abstract

Organisms usually cope with change in the environment by altering the dynamic trajectory of gene expression to adjust the complement of active proteins. The identification of particular sets of genes whose expression is adaptive in response to environmental changes helps to understand the mechanistic base of gene-environment interactions essential for organismic development. We describe a computational framework for clustering the dynamics of gene expression in distinct environments through Gaussian mixture fitting to the expression data measured at a set of discrete time points. We outline a number of quantitative testable hypotheses about the patterns of dynamic gene expression in changing environments and gene-environment interactions causing developmental differentiation. The future directions of gene clustering in terms of incorporations of the latest biological discoveries and statistical innovations are discussed. We provide a set of computational tools that are applicable to modeling and analysis of dynamic gene expression data measured in multiple environments.

Figure 1:

Trajectories of expression for ten genes randomly chosen from those associated with response to vein bypass grafting in rabbits in two treatments, high flow and low flow.

Figure 2:

Plot of BIC values calculated for expression trajectories of different gene clusters over cluster number and LOP order.

Figure 3:

Mean expression trajectories (thick purple curve) of each of the eight gene clusters, labeled (A)–(H), over time under high flow and low flow. The two broken curves in each plot are the 95% confidence intervals of each estimated mean trajectory. Expression trajectories of all those genes that belong to a particular cluster, determined by the posterior probabilities, are shown in light blue.

Figure 4:

Comparison of estimated (solid) and true (dot) mean expression trajectories for each of the eight clusters, labeled (A)–(H), under two hypothesized treatments, high flow and low flow. The simulated data mimicked the structure of the rabbit data, but assuming an error variance that triples the estimated variance. The 95% confidence intervals (broken curves) of each estimated mean trajectory are given. Expression trajectories of all those genes that belong to a particular cluster are shown in light blue.