Institute of Bioinformatics
Center of Informatics,Statistics and Epidemiology UMG, University of Göttingen
This project is dedicated to continuous updating of a previous classification of eukaryotic transcription factors (TFs) according to their DNA-binding domains (1,2). The present classification focuses on human TFs, but will be extended soon to other mammals, and to other eukaryotic taxa in the future. The aim of this classification is to provide a basis for deciphering the protein-DNA recognition code. That would lead to a comprehensive understanding of how the read-out of regulatory information encoded in the genome by proteins works.
By inspection of structural and sequence similarities of the proteins that are known or can be reasonably assumed to act as TFs, a six-level classification scheme was created to which altogether more than 1500 human TFs genes, coding for more than 2900 proteins, can be assigned. They were grouped into 111 families, many of them divided into subfamilies. They constitute 40 classes and 9 superclasses (formerly only 4). Further details are given in a paper for the upcoming Nucleic Acids Research Database Issue (3).
The access to the classification has been organized by J. Dönitz? OBA server (4). The project page can be found here:
The most recent draft version can be found as flat html document on Prof. E. Wingender?s personal home page: