In this issue: Proteomics 10/2008

In this issue of Proteomics you will find the following highlighted articles: Open‐pit mining for compatible neighbors Open pit mines are an explosive topic in some parts of the US and elsewhere around the world. In this case, however, it is information, not coal or copper, that is being mined from computer files. Ahmad et al. are looking for patterns in the sequence of amino acids that surround a landmark, an amino acid that is frequently modified by phosphorylation or glycosylation. If an appropriate set of rules can be found, it becomes feasible to predict sites of post‐translational modification (PTM) and possibly winners in conflicts from overlapping sites. Algorithms (MAPRes) for O‐glycosylation (GalNAc) and O‐phosphorylation have been implemented that show good fit, correlating well with patterns predicted by existing software. The MAPRes software should also be useful in creating patterns for features such as protease targets and secondary protein structures. Ahmad, I. et al., Proteomics 2008, 8 , 1954–1958. Synthetic sequence steals enzyme‐specific (PKCα) spot It is interesting that evolution has optimized, rather than maximized, many interactions. It was only after we maximized these interactions artificially that we began to recognize the subtleties possible with control systems that were not pushed to the max full time. On the other hand, less than 100% is not satisfactory if we are trying to clean out metastasizing tumor cells. Kang et al. are looking for maximum discrimination between protein kinase C (PKC) isozymes for diagnostic and therapeutic applications. PKCα is normally involved in differentiation, growth, and programmed death of many cell types. The researchers began by designing and screening a set of >1700 PKCα target peptides. They selected the one with the highest efficiency of being labeled and characterized it further for kinetics ( K m , and V max ) with 11 PKC isozymes. They also used it for Western blot evaluation of enzyme levels in tumor and normal tissues. Kang, J.‐H. et al., Proteomics 2008, 8 , 2006–2011. Subtleties of B. subtilis biological labeling Bacilllus subtilis is a workhorse bacterium, if you'll allow a mixed metaphor. My grad school friends who worked with it always claimed it was a “higher organism” than E. coli because it could differentiate, sort of like yeast. Because it is well studied genetically and physiologically, it has been adopted as a useful model system for the study of stress responses. Dreisbach et al. wanted to extend proteome analysis to membrane proteins under different starvation conditions that generated the stringent response. Conventional methods ( e.g. 2‐DE) were not quantitative enough, or had unacceptable error rates ( in vitro labelling). They found in vivo labelling with either specific amino acids (SILAC with lysine) or general metabolic labelling (14N/15N‐metabolic) to meet their needs. Samples could be mixed with controls prior to extraction and digestion to markedly reduce technical error rates. Both methods were considered suitable for quantitative proteomic analysis of membrane proteins. Dreisbach, A. et al., Proteomics 2008, 8 , 2062–2076.