The ability to measure the transcriptional response of cells has drawn much attention to the underlying transcriptional networks. To untangle the network, numerous models with corresponding reverse engineering methods have been applied. In this work, we propose a non-linear model with adjustable degrees of complexity. The corresponding reverse engineering method uses a probabilistic scheme to reduce the reconstruction problem to subnetworks. Adequate models for gene regulatory networks must be anchored on sufficient biological knowledge. Here, the cI auto-inhibition circuit (cI circuit) is used to validate our reverse engineering method. Simulations of the cI circuit are used for the reconstruction, whereas a simplified cI circuit model assists the modeling phase. Several levels of complexity are evaluated, subsequently the reconstructed models show different properties. As a result, we reconstruct an abstract model, capturing the dynamic behavior of the cI circuit to a high degree.
@inproceedings{2005_22,
author = {Supper, Jochen and Spieth, Christian and Zell, Andreas},
title = {Reverse Engineering Non-Linear Gene Regulatory Networks Based on
the Bacteriophage lambda cI Circuit},
booktitle = {Proceedings of the 2005 IEEE Symposium on Computational Intelligence
and Computational Biology (CIBCB '05)},
year = {2005},
pages = {325-332},
address = {San Diego, USA},
month = nov,
publisher = {IEEE},
abstract = {The ability to measure the transcriptional response of cells has drawn
much attention to the underlying transcriptional networks. To untangle
the network, numerous models with corresponding reverse engineering
methods have been applied. In this work, we propose a non-linear
model with adjustable degrees of complexity. The corresponding reverse
engineering method uses a probabilistic scheme to reduce the reconstruction
problem to subnetworks. Adequate models for gene regulatory networks
must be anchored on sufficient biological knowledge. Here, the cI
auto-inhibition circuit (cI circuit) is used to validate our reverse
engineering method. Simulations of the cI circuit are used for the
reconstruction, whereas a simplified cI circuit model assists the
modeling phase. Several levels of complexity are evaluated, subsequently
the reconstructed models show different properties. As a result,
we reconstruct an abstract model, capturing the dynamic behavior
of the cI circuit to a high degree.},
doi = {10.1109/CIBCB.2005.1594936},
isbn = {0-7803-9387-2},
url = {http://www.cogsys.cs.uni-tuebingen.de/publikationen/2005/Supper05reveng.pdf}
}