OIPI HBBQ BQGI SKGI DGPIaComparison with recognized motifs Identified motif b-turn b-turn b-turn b-turn b-turn b-turn b-turn b-turn b-turn b-turn b-turn b-turn end b-turn end Nest Nest Nest Nest Nest Nest Nest Niche Niche Matchb b-turn comparison Motivated Proteins comparisonc NA NA Precision Occurrence Lpmaxnbsfnbsfa : nbsf denote the amount of SCOP superfamilies in which a structural word occurs. b: match order ML RR-S2 CDA (ammonium salt) denotes the number of fragments containing a identified motif. c: comparison with Motivated Proteins motifs is restricted towards the set of proteins frequent to our database along with the Motivated Proteins database. Within this case, the number among brackets denotes the number of fragments inved in the comparison.Figure (reduced panel) supplies an instance of a structural word, DRPI, containing a nest motif. We observe that DRPI-fragments are extremely related when it comes to structure and present some weak amino-acid specificities in positions : andThis word 
is recurrent (observed times inside the initial data set and in superfamilies) and over-represented in superfamilies having a Lpmax equal to The representation of two proteins containing the DRPI word shows it really is present in superfamilies with unique folds. Like turn motifs, nest and niche motifs are detected by applying geometrical thresholds. In this case also, the truth that an extremely smaller proportion of our fragments fail the assignment recommend that structural words may very well be used to assign nest- and niche-like motifs.Extreme ubiquitous words not connected to identified structural motifsniche and nest motifs. Let us note, on the other hand, that DGPI is structurally close for the structural word DRPI (RMSD equal towhich consists of nest motifs. Within the similar way, SKGI is related to SLGI (RMSD equal toa word containing the finish of a b-turn.Hyperlink between ubiquitous words and functional annotationsTwo ubiquitous words, DGPI and SKGI, are extracted from proteins not listed in the Motivated Protein database. It is for that reason not attainable to evaluate them inside the earlier component, we’ve shown that intense ubiquitous words include some recognized motifs including turns, nest, niche. It has been shown that these smaller motifs could possibly be inved in protein functions for example active web sites or binding web pages ,. We therefore surveyed the association between extreme ubiquitous words and Swiss-Prot annotation by computing the precision on the extreme ubiquitous words order KRIBB11 toward biological annotations. As reported in More file : Table S, we obtained low precisions, suggesting that ubiquitous words will not be strongly related to functional capabilities.Regad et al. BMC Bioinformatics , : http:biomedcentral-Page ofFigure Illustration of two ubiquitous structural words. Upper aspect: structural word PZCD. Reduced portion: structural word DRPI. A: geometry of numerous word fragments, optimally superimposed. B: amino-acid conservation in the word generated by WebLogo http:weblogo.berkeley. edu. C: word statistics. D: instance of structures containing the structural word. The location of structural word is indicated by arrows.Regad et al. BMC Bioinformatics , : http:biomedcentral-Page ofLink in between intense superfamily-specific words and biological annotationsUnlike ubiquitous words, superfamily-specific words PubMed ID:http://www.ncbi.nlm.nih.gov/pubmed/19387489?dopt=Abstract are highly over-represented in handful of superfamilies, suggesting a feasible implication in function. Within this section, we concentrate our evaluation around the extreme superfamily-specific words, defined by Lpmax and nbsf , and investigate their corresponden.OIPI HBBQ BQGI SKGI DGPIaComparison with recognized motifs Identified motif b-turn b-turn b-turn b-turn b-turn b-turn b-turn b-turn b-turn b-turn b-turn b-turn finish b-turn finish Nest Nest Nest Nest Nest Nest Nest Niche Niche Matchb b-turn comparison Motivated Proteins comparisonc NA NA Precision Occurrence Lpmaxnbsfnbsfa : nbsf denote the amount of SCOP superfamilies in which a structural word occurs. b: match denotes the amount of fragments containing a recognized motif. c: comparison with Motivated Proteins motifs is restricted to the set of proteins popular to our database along with the Motivated Proteins database. In this case, the quantity amongst brackets denotes the number of fragments inved in the comparison.Figure (decrease panel) delivers an instance of a structural word, DRPI, containing a nest motif. We observe that DRPI-fragments are very related when it comes to structure and present some weak amino-acid specificities in positions : andThis word is recurrent (noticed instances within the initial information set and in superfamilies) and over-represented in superfamilies having a Lpmax equal to The representation of two proteins containing the DRPI word shows it can be present in superfamilies with different folds. Like turn motifs, nest and niche motifs are detected by applying geometrical thresholds. Within this case also, the truth that an incredibly small proportion of our fragments fail the assignment suggest that structural words could possibly be made use of to assign nest- and niche-like motifs.Extreme ubiquitous words not connected to recognized structural motifsniche and nest motifs. Let us note, even so, that DGPI is structurally close for the structural word DRPI (RMSD equal towhich contains nest motifs. Within the identical way, SKGI is equivalent to SLGI (RMSD equal toa word containing the finish of a b-turn.Hyperlink in between ubiquitous words and functional annotationsTwo ubiquitous words, DGPI and SKGI, are extracted from proteins not listed within the Motivated Protein database. It is actually consequently not doable to evaluate them within the preceding part, we have shown that extreme ubiquitous words contain some identified motifs for instance turns, nest, niche. It has been shown that these compact motifs could possibly be inved in protein functions for instance active web-sites or binding sites ,. We thus surveyed the association in between extreme ubiquitous words and Swiss-Prot annotation by computing the precision on the intense ubiquitous words toward biological annotations. As reported in Added file : Table S, we obtained low precisions, suggesting that ubiquitous words usually are not strongly associated to functional features.Regad et al. BMC Bioinformatics , : http:biomedcentral-Page ofFigure Illustration of two ubiquitous structural words. Upper element: structural word PZCD. Lower part: structural word DRPI. A: geometry of numerous word fragments, optimally superimposed. B: amino-acid conservation 
on the word generated by WebLogo http:weblogo.berkeley. edu. C: word statistics. D: instance of structures containing the structural word. The location of structural word is indicated by arrows.Regad et al. BMC Bioinformatics , : http:biomedcentral-Page ofLink involving intense superfamily-specific words and biological annotationsUnlike ubiquitous words, superfamily-specific words PubMed ID:http://www.ncbi.nlm.nih.gov/pubmed/19387489?dopt=Abstract are extremely over-represented in handful of superfamilies, suggesting a attainable implication in function. In this section, we concentrate our analysis around the intense superfamily-specific words, defined by Lpmax and nbsf , and investigate their corresponden.