Diagrams of MARK and tau. (A) Top: bar diagram representing the domain structure of MARK family kinases. Residue numbers refer to the sequence of the longest isoform of human MARK2. The domains include the header domain, the kinase catalytic domain, the common docking (CD) site, the ubiquitin-associated (UBA) domain, the spacer domain, and the tail domain, which contains the 'kinase-associated domain' (KA1). Bottom: bar diagram of tau protein, highlighting the repeat domain and the KXGS motifs that can be phosphorylated by MARK. (B) Phylogenetic branch of the AMPK-related kinases and their position in the human kinome (modified from ). Closed squares and circles indicate UBA and KA1 domains recognized on the basis of sequence similarity by Prosite . Open squares and circles label kinases where UBA and KA1 domains are not predicted by sequence, because the homology is too weak, but only on the basis of secondary structure elements (determined using PSIPRED ).
Timm et al. BMC Neuroscience 2008 9(Suppl 2):S9 doi:10.1186/1471-2202-9-S2-S9