Evaluating influence of microRNA in reconstructing gene regulatory networks
- Authors: Chowdhury, Ahsan , Chetty, Madhu , Nguyen, Vinh
- Date: 2015
- Type: Text , Journal article
- Relation: Cognitive neurodynamics Vol. 8, no. 3 (2015), p. 251-9
- Full Text: false
- Reviewed:
- Description: Gene regulatory network (GRN) consists of interactions between transcription factors (TFs) and target genes (TGs). Recently, it has been observed that micro RNAs (miRNAs) play a significant part in genetic interactions. However, current microarray technologies do not capture miRNA expression levels. To overcome this, we propose a new technique to reverse engineer GRN from the available partial microarray data which contains expression levels of TFs and TGs only. Using S-System model, the approach is adapted to cope with the unavailability of information about the expression levels of miRNAs. The versatile Differential Evolutionary algorithm is used for optimization and parameter estimation. Experimental studies on four in silico networks, and a real network of Saccharomyces cerevisiae called IRMA network, show significant improvement compared to traditional S-System approach.
Inferring large scale genetic networks with S-System model
- Authors: Chowdhury, Ahsan , Chetty, Madhu , Nguyen, Vinh
- Date: 2013
- Type: Text , Conference paper
- Relation: Genetic and Evolutionary Computation Conference p. 271-278
- Full Text: false
- Reviewed:
- Description: Gene regulatory network (GRN) reconstruction from high-throughput microarray data is an important problem in systems biology. The S-System model, a differential equation based approach, is among the mainstream approaches for modeling GRNs
On the analysis of time-delayed interactions in genetic network using S-system model
- Authors: Chowdhury, Ahsan , Chetty, Madhu , Nguyen, Vinh
- Date: 2013
- Type: Text , Conference paper
- Relation: 20th International Conference, ICONIP 2013 Daegu, Korea, November 3rd-7th; published in Neural Information Processing, Lecture Notes in Computer Science, vol 8227. pg 616-623
- Full Text: false
- Reviewed:
- Description: The Gene Regulatory Network (GRN) is the collection of genes and interactions among them, which captures the mutual interactions among genes. Amongst the various currently available models for inferring GRN, the S-System formalism is often considered as an excellent compromise between accuracy and mathematical tractability, although limited to represent the instantaneous interactions only. Recently proposed Time-delayed S-System Model (TDSS), an improved version of the traditional S-System model, is capable of representing the delayed interactions present in the genetic network. In this paper, we have shown the results of extensive analysis performed on TDSS over a widely used synthetic network. The two well-known performance measures applied to the synthetic network with various time-delayed regulations clearly demonstrate that the TDSS can capture both the instantaneous and delayed interactions correctly with high precision. Further, we have shown the effect of various samples sizes during the optimization where average error for the inferred parameters are reported and compared with an existing state-or-the-art algorithm. © Springer-Verlag 2013.
Reverse engineering genetic networks with time-delayed S-system model and pearson correlation coefficient
- Authors: Chowdhury, Ahsan , Chetty, Madhu , Nguyen, Vinh
- Date: 2013
- Type: Text , Conference paper
- Relation: 20th International Conference on Neural Information Processing, ICONIP 2013; Daegu; South Korea; 3rd-7th November 2013; published in Neural Information Processing, Lecture Notes in Computer Science, vol 8227. p. 624-631
- Full Text: false
- Reviewed:
- Description: In almost all biological systems including genetic networks, the complex simultaneous interactions occurring amongst different organelles within a cell are both - instantaneous and time-delayed. Among the various modeling approaches, applied for inferring Gene Regulatory Network (GRN), recently proposed Time-delayed S-System Model (TDSS) is capable of simultaneously represent both the instantaneous and time-delayed interactions. While the delay parameters are incorporated in the S-System model to propose TDSS, this open a new challenge in GRN reconstruction. This paper proposes a systematic approach to fit in various level of knowledge in the delay parameters during the reverse engineering process. Further, we have approximated the delay parameters with well-known statistical measure Pearson correlation coefficient. Experimental studies have been carried out considering two widely used synthetic networks with various delays and real-life network of Saccharomyces cerevisiae called IRMA. The results clearly exhibit the influence of incorporating knowledge in the parameter learning process. © Springer-Verlag 2013.
On the reconstruction of genetic network from partial microarray data
- Authors: Chowdhury, Ahsan , Chetty, Madhu , Nguyen, Vinh
- Date: 2012
- Type: Text , Conference paper
- Relation: 19th international conference on Neural Information Processing, ICONIP'12, Doha, Qatar; In Neural Information Processing - Volume Part I (Lecture Notes in Computer Science series) Vol. 7663 , p.689-696 p. 689-696
- Full Text: false
- Reviewed:
- Description: Gene Regulatory Network (GRN) contains interactions occurring between transcription factors (TF) and target genes which are captured during the microarray creation. However, information about the interactions among microRNAs (miRNA) and target genes can not be captured by current microarray technology. To overcome this limitation, we propose a new technique to reverse engineer GRN from partial microarray data which represent target genes' interactions only. Using S-System model, the approach is modified to incorporate the unavailability of information about miRNA-target genes interactions. The most versatile Differential Evolutionary algorithm is used for optimization and parameter learning. Experimental studies on three newly created synthetic networks, and one real network of Saccharomycescerevisiae called IRMA network, show significant improvement compared to traditional S-System based approach.