Identifiability and inference of pathway motifs by epistasis analysis

Abstract

The accuracy of genetic network inference is limited by the assumptions used to determine if one hypothetical model is better than another in explaining experimental observations. Most previous work on epistasis analysis-in which one attempts to infer pathway relationships by determining equivalences among traits following mutations-has been based on Boolean or linear models. Here, we delineate the ultimate limits of epistasis-based inference by systematically surveying all two-gene network motifs and use symbolic algebra with arbitrary regulation functions to examine trait equivalences. Our analysis divides the motifs into equivalence classes, where different genetic perturbations result in indistinguishable experimental outcomes. We demonstrate that this partitioning can reveal important information about network architecture, and show, using simulated data, that it greatly improves the accuracy of genetic network inference methods. Because of the minimal assumptions involved, equivalence partitioning has broad applicability for gene network inference.

The potential power of dynamics in epistasis analysis

“…It would allow the continuation to assess the theoretical value of dynamics in epistasis analysis and the development of algorithms that will be directly applicable to real-world data. The combination of quantitative statistical, regression or probabilisticbased approaches [21,28,29,30] with dynamical models in epistasis analysis could lead to methods with great practical utility in determining genetic pathway structures. Will the assumptions made act as a practical barrier to the understanding of gene networks?…”

“…We also restrict the type of GRNs to only acyclic networks. Although most studies on epistasis analysis assume the acyclicity of gene networks [11][12][13][14][15][16][17][18][19][20][21][22][23][24][25][26][27][28], many biological processes rely on feedback mechanisms for regulation [26]. They may appear as loops in which one gene regulates itself or two or more genes regulate each other.…”

“…Yeast S. cerevisiae has been extensively explored [17,18,20,22,24,25,27,28]. Other studies include growth and development of amoeba Dictyostelium discoideum [19,22] and programmed cell death, sex determination or vulval development in C. elegans [11,13].…”

“…We now examine a classical method from genetics, which in its simplest form can be performed by handepistasis analysis [11,12,13]. Epistasis is a vital tool in functional genomics to enhance our understanding of the GRN components and their order of action [11][12][13][14][15][16][17][18][19][20][21][22][23][24][25][26][27][28][29].…”

“…More specifically, it refers to situations where the outcome of deleting two genes is "surprising" compared to the results of deleting each gene individually. While classical epistasis has yielded deeper insights on numerous genetic pathways [11][12][13][14][15][16][17][18][19][20][21][22][23][24][25][26][27][28][29], it is not without limitations. One limitation is that it is not always possible to infer the relationship between a pair of genes using epistasis analysis.…”