Inferring genetic interactions via a nonlinear model and an optimization algorithm

Abstract

BackgroundBiochemical pathways are gradually becoming recognized as central to complex human diseases and recently genetic/transcriptional interactions have been shown to be able to predict partial pathways. With the abundant information made available by microarray gene expression data (MGED), nonlinear modeling of these interactions is now feasible. Two of the latest advances in nonlinear modeling used sigmoid models to depict transcriptional interaction of a transcription factor (TF) for a target gene, but do not model cooperative or competitive interactions of several TFs for a target.ResultsAn S-shape model and an optimization algorithm (GASA) were developed to infer genetic interactions/transcriptional regulation of several genes simultaneously using MGED. GASA consists of a genetic algorithm (GA) and a simulated annealing (SA) algorithm, which is enhanced by a steepest gradient descent algorithm to avoid being trapped in local minimum. Using simulated data with various degrees of noise, we studied how GASA with two model selection criteria and two search spaces performed. Furthermore, GASA was shown to outperform network component analysis, the time series network inference algorithm (TSNI), GA with regular GA (GAGA) and GA with regular SA. Two applications are demonstrated. First, GASA is applied to infer a subnetwork of human T-cell apoptosis. Several of the predicted interactions are supported by the literature. Second, GASA was applied to infer the transcriptional factors of 34 cell cycle regulated targets in S. cerevisiae, and GASA performed better than one of the latest advances in nonlinear modeling, GAGA and TSNI. Moreover, GASA is able to predict multiple transcription factors for certain targets, and these results coincide with experiments confirmed data in YEASTRACT.ConclusionsGASA is shown to infer both genetic interactions and transcriptional regulatory interactions well. In particular, GASA seems able to characterize the nonlinear mechanism of transcriptional regulatory interactions (TIs) in yeast, and may be applied to infer TIs in other organisms. The predicted genetic interactions of a subnetwork of human T-cell apoptosis coincide with existing partial pathways, suggesting the potential of GASA on inferring biochemical pathways.

Modeling the Zebrafish Segmentation Clock’s Gene Regulatory Network Constrained by Expression Data Suggests Evolutionary Transitions Between Oscillating and Nonoscillating Transcription

“…Simulated annealing has been used previously to infer GRNs from microarray data sets and fit multiple parameter values (Tomshine and Kaznessis 2006;Chen et al 2010). However, these studies focused only on in silico or synthetic networks where the underlying topology was already known.…”

Modeling nonlinear dynamic biological systems with human-readable fuzzy rules optimized by convergent heterogeneous particle swarm

“…Cao, Rees, and Feng (1997), fuzzy clustering method is employed to identify T-S fuzzy models, including identification of the number of fuzzy rules and parameters of fuzzy membership functions. In traditional methods, the structures and the parameters of the models are determined separately in different stages described above as opposed to simultaneously, where the structure of a model is first derived by S-system or neural network (Almeida & Voit, 2003;Kikuchi, Tominaga, Arita, Takahashi, & Tomita, 2003;Xu, Wunsch, & Frank, 2007), and then the associated parameters are optimized by evolutionary algorithms (Tsai & Wang, 2005) or heuristic methods (Chen, Lee, Chuang, Wang, & Shieh, 2010;Sun, Garibaldi, & Hodgman, 2012;Xu et al, 2007). However, it is difficult to guarantee that the obtained models will have good performance because of the potential inconsistence between the two steps.…”

Seeding-inspired chemotaxis genetic algorithm for the inference of biological systems

“…Various variants of differential evolutions and GAs have been widely used (Noman and Iba, 2005;Tsai and Wang, 2005;Wu et al, 2006a,b;Liu and Wang, 2009;Wang and Liu, 2010;Kikuchi et al, 2003;Voit and Almeida, 2004;Ho et al, 2007). Other approaches such as simulated annealing (SA) (Gonzalez et al, 2007), a hybrid of SA and GA (Chen et al, 2010), radial basis functions (Matsubara et al, 2006) and neural networks (Murata et al, 2008) have been proposed to achieve topdown modeling from time series data.…”