ayout 1 - EMBL Grenoble

Structural and Computational Biology UnitBiological sequence analysisPrevious and current researchThe group seeks to gain insight through the computational analysis of biological molecules, particularlyat the protein sequence level. To this end, we deploy many sequence analysis methods andlook to develop new tools as the need arises. Where possible, we contribute to multidisciplinaryprojects involving structural and experimental groups at EMBL and elsewhere. We are probablybest known for our involvement with the Clustal W and Clustal X programs that are widely usedfor multiple sequence alignment, working closely with Julie Thompson (Strasbourg) and Des Higgins(Dublin) to maintain and develop these programs. We also maintain several public web serversat EMBL, including ELM, the protein linear motif resource; Phospho.ELM, a collection of >18,000reported phosphorylation sites; and GlobPlot, a tool for exploring protein disorder.A major focus recently has been to develop and deploy tools for protein architecture analysis. Ourgroup coordinated the EU-funded ELM consortium that developed the Eukaryotic Linear Motifresource to help users find functional sites in modular protein sequences. Short functional sites(e.g. figure 1) are used for the dynamic regulation of large cellular protein complexes and theircharacterisation is essential for understanding cell signalling. So-called ‘hub’ proteins that makemany contacts in interaction networks are thought to have abundant regulatory motifsin large segments of IUP (intrinsically unstructured protein segments). Freelyavailable ELM resource data is now used by many bioinformatics groups to improveprediction of linear motif interactions, e.g. the NetworKIN kinase-substrate predictorand the DILIMOT and SLIMFinder novel motif predictors.Toby GibsonPhD 198, CambridgeUniversity.Postdoctoral research at theLaboratory of MolecularBiology, Cambridge.Team leader at EMBL since1986.Future projects and goalsComputers are applied in molecular biology in the hope that, ultimately, they willinform experimental strategies. As an example, we have recently proposed new candidateKEN boxes, a sequence motif that targets cell cycle proteins for destruction inanaphase (figure 2). We will continue to survey individual gene families in depth andwill undertake proteome surveys when we have specific questionsto answer. Molecular evolution is one of the group’s interests, especiallywhen it has practical applications.With our collaborators, we will look to build up the protein architecturetools, especially the unique ELM resource, taking themto a new level of power and applicability. We are currently workingto add structure and conservation filtering to ELM. We willapply the tools in the investigation of modular protein functionand may deploy them in proteome and protein network analysispipelines. Our links to experimental and structural groups shouldensure that bioinformatics results feed into experimental analysesof signalling interactions and descriptions of the structures ofmodular proteins and their complexes, with one focus being regulatorychromatin proteins.Figure 1 (above): Structure of a typical linear motif-ligand domaininteraction. Here the Rad9 FHA domain is bound to a phosphothreoninepeptide (pdb:1K3N). Annotated in ELM as LIG_FHA_2.Figure 2: A candidate KEN box in the important cell cyclekinase Hipk2. The sequence segment is predicted to benatively disordered and has many conserved phosphorylationmotifs as well as the KEN motif. (Michael et al., 2008)Selected referencesDiella, F., Gould, C.M., Chica, C., Via, A. & Gibson, T.J. (2008).Phospho.ELM: a database of phosphorylation sites – update 2008.Nucleic Acids Res., 36, D20-D2Diella, F., Haslam, N., Chica, C., Budd, A., Michael, S., Brown, N.P.,Trave, G. & Gibson, T.J. (2008). Understanding eukaryotic linearmotifs and their role in cell signaling and regulation. Front Biosci., 13,6580-6603Michael, S., Trave, G., Ramu, C., Chica, C. & Gibson, T.J. (2008).Discovery of candidate KEN-box motifs using cell cycle keywordenrichment combined with native disorder prediction and motifconservation. Bioinformatics, 2, 53-57Perrodou, E., Chica, C., Poch, O., Gibson, T.J. & Thompson, J.D.(2008). A new protein linear motif benchmark for multiple sequencealignment software. BMC Bioinformatics, 9, 2137