×

Second order optimization for the inference of gene regulatory pathways. (English) Zbl 1296.92146

Summary: With the increasing availability of experimental data on gene interactions, modeling of gene regulatory pathways has gained special attention. Gradient descent algorithms have been widely used for regression and classification applications. Unfortunately, results obtained after training a model by gradient descent are often highly variable. In this paper, we present a new second order learning rule based on the Newton’s method for inferring optimal gene regulatory pathways. Unlike the gradient descent method, the proposed optimization rule is independent of the learning parameter. The flow vectors are estimated based on biomass conservation. A set of constraints is formulated incorporating weighting coefficients. The method calculates the maximal expression of the target gene starting from a given initial gene through these weighting coefficients. Our algorithm has been benchmarked and validated on certain types of functions and on some gene regulatory networks, gathered from literature. The proposed method has been found to perform better than the gradient descent learning. Extensive performance comparison with the extreme pathway analysis method has underlined the effectiveness of our proposed methodology.

MSC:

92D10 Genetics and epigenetics

Software:

C-ImmSim
PDFBibTeX XMLCite
Full Text: DOI

References:

[1] Angenent, G. C., J. Franken, M. Busscher, L. Colombo and A. J. van Tunen (1993): “Petal and stamen formation in petunia is regulated by the homeotic gene fbp1,” Plant J., 4, 101-112.;
[2] Aubert, D., L. Chen, Y. Moon, D. Martin, L. A. Castle, C. Yang and Z. R. Sung (2001): “Emf1, a novel protein involved in the control of shoot architecture and flowering in arabidopsis,” Plant Cell, 13, 1865-1875.;
[3] Ay, F., F. Xu and T. Kahveci (2009): “Scalable steady state analysis of boolean biological regulatory networks,” Plos One, 4, e7992, doi: 10.1371/journal.pone.0007992.;
[4] Bahler, J. (2005): “Cell-cycle control of gene expression in budding and fission yeast,” Annu. Rev. Genet., 39, 69-94.;
[5] Barrett, C. L. and B. O. Palsson (2006): “Iterative reconstruction of transcriptional regulatory networks: an algorithmic approach,” Plos Comput. Biol., 2, e52, doi: 10.1371/journal.pcbi.0020052.;
[6] Benyamini, T., O. Folger, E. Ruppin and T. Shlomi (2010): “Flux balance analysis accounting for metabolite dilution,” Genome Biol., 11, R43, doi: 10.1186/gb-2010-11-4-r43.;
[7] Bortoletti, A., C. D. Fiore, S. Fanelli and P. Zellini (2003): “A new class of quasi-newtonian methods for optimal learning in mlp-networks,” IEEE T. Neural Network, 14, 263-273.;
[8] Bowman, J. L., J. Alvarez, D. Weigel, E. M. Meyerowitz and D. R. Smyth (1993): “Control of flower development in arabidopsis thaliana by apetala1 and interacting genes,” Development, 119, 721-743.;
[9] Buntine, W. L. and S. Weigend (1994): “Computing second derivatives in feed-forward networks: a review,” IEEE T. Neural Network, 5, 480-488.;
[10] Calonje, M., R. Sanchez, L. Chen and Z. R. Sung (2008): “Embryonic flower1 participates in polycomb group-mediated ag gene silencing in arabidopsis,” Plant Cell, 20, 277-291.;
[11] Castillo, E., B. G. Berdinas, O. F. Romero and A. A. Betanzos (2006): “A very fast learning method for neural networks based on sensitivity analysis,” J. Mach. Learn. Res., 7, 1159-1182.; · Zbl 1222.68159
[12] Coen, E. and E. Meyerowitz (1991): “The war of the whorls: genetic interactions flower development,” Nature, 353, 31-37.;
[13] Coupland, G. (1995): “Genetic and enviromental control of flowering time in arabidopsis,” Trends Genet., 11, 393-397.;
[14] Crombach, A. and P. Hogeweg (2008): “Evolution of evolvability in gene regulatory networks,” PLoS Comput. Biol., 4, e1000112, doi: 10.1371/journal.pcbi.1000112.;
[15] Das, M., S. Mukhopadhyay and R. K. De (2010): “Gradient descent optimization in gene regulatory pathways,” PLoS One, 5, e12475, doi: 10.1371/journal.pone.0012475.;
[16] Davidich, M. I. and S. Bornholdt (2008): “Boolean network model predicts cell cycle sequence of fission yeast,” PLoS One, 3, e1672, doi: 10.1371/journal.pone.0001672.; · Zbl 1400.92207
[17] Garg, A., I. Xenarios, L. Mendoza and G. DeMicheli (2007): “Efficient methods for dynamic analysis of genetic networks and in silico gene perturbation experiments,” Lect. Notes Bioinform., 4453, 62-76.;
[18] Goelzer, A., F. B. Brikci, I. M. Verstraete and P. e. a. Noirot (2008): “Reconstruction and analysis of the genetic and metabolic regulatory networks of the central metabolism of bacillus subtilis,” BMC Syst. Biol., 2, 20, doi: 10.1186/1752-0509-2-20.;
[19] Hong, T., J. Xing, L. Li and J. J. Tyson (2011): “A mathematical model for the reciprocal differentiation of t helper 17 cells and induced regulatory t cells,” Plos Comput. Biol., 7, e1002122, doi: 10.1371/journal.pcbi.1002122.;
[20] Ingolia, N. T. (2004): “Topology and robustness in the Drosophila segment polarity network,” PLoS Biol., 2, e123, doi: 10.1371/journal.pbio.0020123.;
[21] Lee, W. and W. Tzou (2009): “Computational methods for discovering gene networks from expression data,” Brief. Bioinform., 10, 408-423.;
[22] Ma, W., L. Lai, Q. Ouyang and C. Tang (2006): “Robustness and modular design of the Drosophila segment polarity network,” Mol. Syst. Biol., 2, 70, doi: 10.1038/msb4100111.;
[23] Magoulas, G. D., M. N. Vrahatis and G. S. Androulakis (1999): “Improving the convergence of the backpropagation algorithm using learning rate adaptation methods,” Neural Comput., 11, 1769-1796.;
[24] Mendoza, L. and E. R. Alvarez-Buylla (1998): “Dynamics of the genetic regulatory network for arabidopsis thalina flower morphogenesis,” J. Theor. Biol., 193, 307-319.;
[25] Naldi, A., J. Carneiro, C. Chaouiya and D. Thieffry (2010): “Diversity and plasticity of Th cell types predicted from regulatory network modelling,” Plos Comput. Biol., 6, e1000912, doi: 10.1371/journal.pcbi.1000912.;
[26] Nanfack, Y. F., M. Postma and J. A. Kaandorp (2009): “Inferring drosophila gap gene regulatory network: a parameter sensitivity and perturbation analysis,” BMC Syst. Biol., 3, 94, doi: 10.1186/1752-0509-3-94.;
[27] Pedicini, M., F. Barrenas, T. Clancy and F. e. a. Castiglione (2010): “Combining network modeling and gene expression microarray analysis to explore the dynamics of Th1 and Th2 cell regulation,” Plos Comput. Biol., 6, e1001032, doi: 10.1371/journal.pcbi.1001032.;
[28] Pnueli, L., D. Hareven, S. D. Rounsley, M. F. Yanofsky and E. Lifshitz (1994): “Isolation of the tomato agamous gene tag1 and analysis of its homeotic role in transgenic plants,” Plant Cell, 6, 163-173.;
[29] Rodriguez, J. S., L. Simeoni, J. A. Lindquist and R. e. a. Hemenway (2007): “A logical model provides insights into T cell receptor signaling,” Plos Comput. Biol., 3, 163, doi: 10.1371/journal.pcbi.0030163.;
[30] Samach, A., J. E. Klenz, S. E. Kohalmi, E. Risseeuw, G.W. Haughn and W. L. Crosby (1999): “The unusual floral organs gene of arabidopsis thaliana is an f-box protein required for normal patterning and growth in the floral meristem,” Plant J., 20, 433-445.;
[31] Sanchez, L., C. Chaouiya and D. Thieffry (2008): “Segmenting the fly embryo: logical analysis of the role of the segment polarity cross-regulatory module,” Int. J. Dev. Biol., 52, 1059-1075.;
[32] Santoni, D., M. Pedicini and F. Castiglione (2008): “Implementation of a regulatory gene network to simulate the th1/2 differentiation in an agent-based model of hypersensitivity reactions,” Bioinformatics, 24, 1374-1380.;
[33] Schilling, C. H., D. Letscher and B. O. Palsson (2000): “Theory for the systemic defnition of metabolic pathways and their use in interpreting metabolic function from a pathway-oriented perspective,” J. Theor. Biol., 203, 229-248.;
[34] Schlitt, T. and A. Brazma (2007): “Current approaches to gene regulatory network modelling,” BMC Bioinformatics, 8 (Suppl 6), S9, doi: 10.1186/1471-2105-8-S6-S9.;
[35] Simpson, G., A. Gendall and C. Dean (1999): “When to switch to flowering,” Annu. Rev. Cell Dev. Biol., 99, 519-550.;
[36] Sirbu, A., H. J. Ruskin and M. Crane (2010): “Comparison of evolutionary algorithms in gene regulatory network model inference,” BMC Bioinformatics, 11, 59, doi: 10.1186/1471-2105-11-59.;
[37] Stelling, J. (2004): “Mathematical models in microbial systems biology,” Curr. Opin. Microbiol., 7, 513-518.;
[38] Sundstrom, J. F., N. Nakayama, K. Glimelius and V. F. Irish (2006): “Direct regulation of the floral homeotic apetala1 gene by apetala3 and pistillata in arabidopsis,” Plant J., 46, 593-600.;
[39] Tu, Z., L. Wang, M. N. Arbeitman, T. Chen and F. Sun (2006): “An integrative approach for causal gene identification and gene regulatory pathway inference,” Bioinformatics, 22, 489-496.;
[40] Wang, Y. J. and C. T. Lin (1998): “A second-order learning algorithm for multilayer networks based on block hessian matrix,” Neural Networks, 11, 1607-1622.;
[41] Xiong, M., J. Zhao and H. Xiong (2004a): “Network-based regulatory pathways analysis,” Bioinformatics, 20, 2056-2066.;
[42] Xiong, M. M., J. Li and X. Z. Fang (2004b): “Identification of genetic networks,” Genetics, 166, 1037-1052.;
[43] Yeh, I. C. and W. L. Cheng (2010): “First and second order sensitivity analysis of mlp,” Neurocomputing, 73, 2225-2233.;
[44] Zhang, X., A. J. Cal and J. O. Borevitz (2011): “Genetic architecture of regulatory variation in arabidopsis thaliana,” Genome Res., 21, 725-733.;
This reference list is based on information provided by the publisher or from digital mathematics libraries. Its items are heuristically matched to zbMATH identifiers and may contain data conversion errors. In some cases that data have been complemented/enhanced by data from zbMATH Open. This attempts to reflect the references listed in the original paper as accurately as possible without claiming completeness or a perfect matching.