These proteins were first identified by similarity to Hidden Markov Models as described below

suggested that P. infestans might contain a TK based on the clustering of an ESTderived sequence with TKs, albeit with weak branch support. Our PubMed ID:http://www.ncbi.nlm.nih.gov/pubmed/19798435 analysis of the full-length gene suggests it is a serine/threonine kinase, however. It therefore appears that TKs predated the radiation of eukaryotic life forms, although convergent evolution can not be excluded. The extent of the importance of tyrosine phosphorylation in oomycetes remains to be elucidated. Only four proteins in P. infestans contain plausible SH2 phosphotyrosine-binding domains and none have PTB phosphotyrosine-binding domains. Dual-specificity phosphatases have been detected, however, which are an indicator of phosphotyrosine signaling. OTHER family This comprises ePKs that do not fit well into the above-described groups, but which nonetheless are typically well-conserved in eukaryotes. P. infestans contains 35 OTHER kinases. These include ePKs involved in cell division such as four Aurora, six NEK, two POLO-like, and two WEE kinases. Also detected were four NAK kinases which in other species regulate the cytoskeleton, one VPS15 kinase which participates in protein sorting, a WNK kinase which regulates ion homeostasis, and five PEK/GCN2 kinases which control translation initiation and participate in the starvation response. Two ULK kinases were also detected, which in yeast control autophagy. Some workers suggested that most of these kinases can be placed in the main RS-1 families. We have not chosen that Judelson and Ah-Fong BMC Genomics 2010, 11:700 http://www.biomedcentral.com/1471-2164/11/700 Page 14 of 20 approach in this study since most group with each other in phylogenetic analysis. The exceptions are Aurora and POLO, which have affinity to the CAMK family. Expression pattern of ePKs Microarray and qRT-PCR analysis was used to measure mRNA levels in hyphae, asexual sporangia, and swimming zoospores to obtain more insight into the function of the kinases. By mining data from our prior microarray study, reliable signals were obtained for 221 ePK genes in one or more of the three developmental stages. The remainder were either not represented or gave poor signals on those microarrays, and to obtain data from these qRT-PCR was performed. In total, expression could be measured reliably for 293 kinases in the three developmental stages by combining both approaches. Five genes were studied using both methods, which revealed similar patterns of expression. A compilation of the data is presented graphically in encodes a NEK kinase, which in other species regulates the formation and stability of the axonemal microtubules of flagella. Perhaps this protein also plays such a role in sporangia, which already contain all proteins needed for zoosporogenesis. Other kinases with intriguing potential functions can be extracted from the data but are too numerous to describe here. Besides suggesting how kinases regulate development, the expression data can help indicate whether the ePK gene models actually lead to a protein; in any genome project, not all predicted proteins are necessarily expressed. Based on data from the microarray and qRTPCR studies, and searches of databases of P. infestans expressed sequence tags for matches to the gene models, it appears that at least 327 of the 354 P. infestans ePK genes are transcribed. Evidence of expression was also obtained for 27 of the 32 predicted catalytically inactive kinases. Judelson and Ah-Fong BMC Genomics 2010, 11:700 http://www.biomedcentr