| Hi-index | 0.00 |
The development of algorithms for reverse-engineering gene regulatory networks is boosted by microarray technologies, which enable the simultaneous measurement of all RNA transcripts in a cell. Meanwhile the curated repository of regulatory associations between transcription factors (TF) and target genes is available based on bibliographic references. In this paper we propose a novel method to combine time-course microarray dataset and documented or potential known transcription regulators for inferring gene regulatory networks. The gene network reconstruction algorithm is based on linear programming and performed in the supervised learning framework. We have tested the new method using both simulated data and experimental data. The result demonstrates the effectiveness of our method which significantly alleviates the problem of data scarcity and remarkably improves the reliability.