Structure-Based Target Prediction of Transcription Factors

Regulation of gene expression in higher organisms is achieved by a complex network of transcription factors and their target genes. Explosive amount of sequence information of genes and transcription factors from genome analyses are presenting a great challenge to bioinformatics. Finding target genes for transcription factors at the genome level will lay a basis for the analysis of gene regulatory network. Sequences contain rich amount of biological information, and comparison of known binding sequences is among the most commonly used methods for the target prediction. However, the accuracy of this method is rather limited, dependent on the quality of sequence information used. Furthermore, because transcription factors usually bind to multiple target sequences and regulate multiple genes in a cooperative manner, the target prediction is a rather complicated problem. In order to tackle such problem, we need to utilize as much information as possible. Structural data on protein-DNA complex contain valuable functional information as well. Due to the progress of X-ray crystallography and NMR spectroscopy techniques, structural data on the protein-DNA complex have been rapidly increasing and more than 200 complexes have been registered in Protein Data Bank. Here, we show that these structural data can be used for the target prediction. Because the structural information is independent of sequence information, it can complement the sequence-based method.