A prior knowledge based approach to infer gene regulatory networks

Authors:
Md. Mahmudul Hasan;Nasimul Noman;Hitoshi Iba
Affiliations:
University of Dhaka, Dhaka, Bangladesh;University of Dhaka, Dhaka, Bangladesh;University of Tokyo, Tokyo, Japan
Venue:
ISB '10 Proceedings of the International Symposium on Biocomputing
Year:
2010

Citing 6
Cited 0

Differential Evolution – A Simple and Efficient Heuristic for Global Optimization over Continuous Spaces

Journal of Global Optimization
Decoupling dynamical systems for pathway identification from metabolic profiles

Bioinformatics
Inference of S-system models of genetic networks using a cooperative coevolutionary algorithm

Bioinformatics
Identification of biochemical networks by S-tree based genetic programming

Bioinformatics
Inferring Gene Regulatory Networks using Differential Evolution with Local Search Heuristics

IEEE/ACM Transactions on Computational Biology and Bioinformatics (TCBB)
Inference of biochemical network models in S-system using multiobjective optimization approach

Bioinformatics

Quantified Score

Hi-index	0.00

Visualization

Abstract

In this research, we use S-System model and Differential Evolution based inference method to capture cellular dynamics using available mutual interaction information among genes. We propose a new fitness function, effectively incorporating a priori information, which guides the inference method to deduce correct skeletal structure of the network with more accurate parameter values. Proposed fitness function mirrors user's confidence in the validity of knowledge and helps in narrowing down the search range of the model parameters for highly confident knowledge. We investigate the potency of the method in terms of quality of data and required data size. The proposed method is shown to perform better in inherent noisy data and in presence of small number of time-dynamics data. We also investigate how the inference method performs in terms of iterative incorporation of knowledge. In inferring cell-cycle data of budding yeast (Saccharomyces cerevisiae), guided by knowledge, the inference method predicts 17 and 23 correct regulations in first and second iteration, respectively which is significantly higher than some other existing methods. Along with finding the parameter values more accurately, it predicts some new regulations and helps in revealing the underlying network structure.