ProDom: automated clustering of homologous domains

The ProDom database is a comprehensive set of protein domain families automatically generated from the SWISS-PROT and TrEMBL sequence databases. An associated database, ProDom-CG, has been derived as a restriction of ProDom to completely sequenced genomes. The ProDom construction method is based on iterative PSI-BLAST searches and multiple alignments are generated for each domain family. The ProDom web server provides the user with a set of tools to visualise multiple alignments, phylogenetic trees and domain architectures of proteins, as well as a BLAST-based server to analyse new sequences for homologous domains. The comprehensive nature of ProDom makes it particularly useful to help sustain the growth of InterPro.     

Unblocked statistical-coil tetrapeptides in aqueous solution: quantum-chemical computation of the carbon-13 NMR chemical shifts

We recently reported a theoretical characterization of representative ensembles of statistical-coil conformations for tetrapeptides with unblocked termini in aqueous solution, at pH 7. The results showed good agreement between the computed Boltzmann-averaged and experimentally-determined values for both the vicinal coupling constants ³J_NH and the -proton chemical shifts. Here, we carry out a cluster analysis of the ensembles of conformations generated in that study, and use them to compute the Boltzmann-averaged values of the quantum-chemical ¹³C chemical shifts for different amino acids in the unblocked tetrapeptides GGXA (where X stands for Phe, Arg, His, Glu, Ile, Lys, Gln, Tyr, Leu, Thr, Ala, Gly and Val). The values of the ¹³C chemical shifts in these thirteen amino acids (for which experimental data are available) were computed by using Density Functional Theory with a 6–311+G(2d,p) basis set. Good agreement is found in terms of both the correlation coefficient (R) and standard deviations of the difference between the computed Bolztmann-averaged and the NMR-determined values for the ¹³C chemical shifts. These results suggest that it may be possible to build a reliable theoretically-derived database of chemical shifts for statistical-coil residues. The results of the current study contribute to our understanding of the relations between chemical shifts, dihedral angles and vicinal coupling constants, ³J_NH. In addition, they can shed light as to how the statistical-coilconformation is related to the conformational preference of more structured states, such as the -helical conformation.     

Selective uncoupling of p120(ctn) from E-cadherin disrupts strong adhesion

p120(ctn) is a catenin whose direct binding to the juxtamembrane domain of classical cadherins suggests a role in regulating cell-cell adhesion. The juxtamembrane domain has been implicated in a variety of roles including cadherin clustering, cell motility, and neuronal outgrowth, raising the possibility that p120 mediates these activities. We have generated minimal mutations in this region that uncouple the E-cadherin-p120 interaction, but do not affect interactions with other catenins. By stable transfection into E-cadherin-deficient cell lines, we show that cadherins are both necessary and sufficient for recruitment of p120 to junctions. Detergent-free subcellular fractionation studies indicated that, in contrast to previous reports, the stoichiometry of the interaction is extremely high. Unlike alpha- and beta-catenins, p120 was metabolically stable in cadherin-deficient cells, and was present at high levels in the cytoplasm. Analysis of cells expressing E-cadherin mutant constructs indicated that p120 is required for the E-cadherin-mediated transition from weak to strong adhesion. In aggregation assays, cells expressing p120-uncoupled E-cadherin formed only weak cell aggregates, which immediately dispersed into single cells upon pipetting. As an apparent consequence, the actin cytoskeleton failed to insert properly into peripheral E-cadherin plaques, resulting in the inability to form a continuous circumferential ring around cell colonies. Our data suggest that p120 directly or indirectly regulates the E-cadherin-mediated transition to tight cell-cell adhesion, possibly blocking subsequent events necessary for reorganization of the actin cytoskeleton and compaction.     

Towards a biogeographic regionalization of the European biota

Aim To determine if it is possible to generate analytically derived regionalizations for multiple groups of European plants and animals and to explore potential influences on the regions for each taxonomic group. Location Europe. Methods We subjected range maps of trees, butterflies, reptiles, amphibians, birds and mammals to k‐means clustering followed by v‐fold cross‐validation to determine the pattern and number of regions (clusters). We then used the mean range sizes of species in each group as a correlate of the number of regions obtained for each taxon, and climate and species richness gradients as correlates of the spatial arrangement of the group‐specific regions. We also included the pattern of tree clusters as a predictor of animal clusters in order to test the ‘habitat templet’ concept as an explanation of animal distribution patterns. Results Spatially coherent clusters were found for all groups. The number of regions ranged from three to eight and was strongly associated with the mean range sizes of the species in each taxon. The cluster patterns of all groups were associated with various combinations of climate, underlying species richness gradients and, in the case of animals, the arrangement of tree clusters, although the rankings of the correlates differed among groups. In four of five groups the tree pattern was the strongest single predictor of the animal cluster patterns. Main conclusions Despite a long history of human disturbance and habitat modification, the European biota retains a discernable biogeographic structure. The primary driver appears to be aspects of climate related to water–energy balance, which also influence richness gradients. For many animals, the underlying habitat structure, as measured by tree distributions, appears to have a strong influence on their biogeographic structure, highlighting the need to preserve natural forest formations if we want to preserve the historical signal found in geographic distributions.     

The management of vegetation classifications with fuzzy clustering

Questions: Does fuzzy clustering provide an appropriate numerical framework to manage vegetation classifications? What is the best fuzzy clustering method to achieve this? Material: We used 531 relevés from Catalonia (Spain), belonging to two syntaxonomic alliances of mesophytic and xerophytic montane pastures, and originally classified by experts into nine and 13 associations, respectively. Methods: We compared the performance of fuzzy C‐means (FCM), noise clustering (NC) and possibilistic C‐means (PCM) on four different management tasks: (1) assigning new relevé data to existing types; (2) updating types incorporating new data; (3) defining new types with unclassified relevés; and (4) reviewing traditional vegetation classifications. Results: As fuzzy classifiers, FCM fails to indicate when a given relevé does not belong to any of the existing types; NC might leave too many relevés unclassified; and PCM membership values cannot be compared. As unsupervised clustering methods, FCM is more sensitive than NC to transitional relevés and therefore produces fuzzier classifications. PCM looks for dense regions in the space of species composition, but these are scarce when vegetation data contain many transitional relevés. Conclusions: All three models have advantages and disadvantages, although the NC model may be a good compromise between the restricted FCM model and the robust but impractical PCM model. In our opinion, fuzzy clustering might provide a suitable framework to manage vegetation classifications using a consistent operational definition of vegetation type. Regardless of the framework chosen, national/regional vegetation classification panels should promote methodological standards for classification practices with numerical tools.     

How different from random are docking predictions when ranked by scoring functions?

Docking algorithms predict the structure of protein–protein interactions. They sample the orientation of two unbound proteins to produce various predictions about their interactions, followed by a scoring step to rank the predictions. We present a statistical assessment of scoring functions used to rank near‐native orientations, applying our statistical analysis to a benchmark dataset of decoys of protein–protein complexes and assessing the statistical significance of the outcome in the Critical Assessment of PRedicted Interactions (CAPRI) scoring experiment. A P value was assigned that depended on the number of near‐native structures in the sampling. We studied the effect of filtering out redundant structures and tested the use of pair‐potentials derived using ZDock and ZRank. Our results show that for many targets, it is not possible to determine when a successful reranking performed by scoring functions results merely from random choice. This analysis reveals that changes should be made in the design of the CAPRI scoring experiment. We propose including the statistical assessment in this experiment either at the preprocessing or the evaluation step. Proteins 2010. © 2010 Wiley‐Liss, Inc.     

The Intrasectional Chemotaxonomic Placement of Hypericum elegans Stephan ex Willd. Inferred from the Essential‐Oil Chemical Composition

Here we report, for the first time, the results of detailed GC and GC/MS analyses of the essential oil of a rare taxon in Serbia, Hypericum elegans Stephan ex Willd . One hundred and sixty two constituents identified accounted for 98.6% of the oil. The major components of the oil were undecane (31.9%), α‐pinene (16.7%), nonane (6.1%), bicyclogermacrene (5.8%), 2‐methyloctane (3.7%), and germacrene D (3.6%). Non‐terpenoids as chemotaxonomic markers constituted the main fraction of H. elegans oil, whereby n‐alkanes were the most abundant contributors of this fraction. Based on these results and previously published ones, we performed an intrasectional multivariate statistical comparison of corresponding essential‐oil chemical compositions. Principal component analysis (PCA) and agglomerative hierarchical clustering (AHC) of the data on the volatile profiles of section Hypericum taxa revealed that H. elegans either represents an oil chemotype of its own (AHC) or could be considered related to H. perforatum (PCA).     

Pairwise Variable Selection for High‐Dimensional Model‐Based Clustering

Summary Variable selection for clustering is an important and challenging problem in high‐dimensional data analysis. Existing variable selection methods for model‐based clustering select informative variables in a “one‐in‐all‐out” manner; that is, a variable is selected if at least one pair of clusters is separable by this variable and removed if it cannot separate any of the clusters. In many applications, however, it is of interest to further establish exactly which clusters are separable by each informative variable. To address this question, we propose a pairwise variable selection method for high‐dimensional model‐based clustering. The method is based on a new pairwise penalty. Results on simulated and real data show that the new method performs better than alternative approaches that use ?₁ and ?_∞ penalties and offers better interpretation.     

Ecological biogeography of Malagasy non‐volant mammals: community structure is correlated with habitat

Aim To examine the relationship between ecoregions, as a proxy for regional climate and habitat type, and mammalian community structure, defined by species composition and richness (e.g. taxonomic structure) and ecological diversity (e.g. ecological structure) of non‐volant species. Location Madagascar. Methods Faunal lists of non‐volant mammal species occurring in 35 communities from five World Wildlife Fund ecoregions were collected from published and unpublished sources. Species were assigned to ecological groups defined by trophic status, locomotor habits, activity cycle and body mass. We used Mantel tests, cluster analysis and principal coordinates analysis to evaluate geographic patterning in taxonomic composition and species richness. We used stepwise multiple discriminant analysis to characterize patterns in the ecological diversity of the mammalian communities from each ecoregion. Communities from transitional habitats (e.g. representing more than one ecoregion) were used to test the predictive power of the analyses. Results Non‐volant mammal communities divided into clusters that correspond to ecoregions. There was a strong distance effect in the taxonomic structure of communities across the island and within both humid and dry forest communities, but this effect was weak within humid forest communities. Mammalian species richness was significantly lower in dry forest than in humid forest communities. The ecological structure of communities was also correlated with ecoregions. Changes in the relative percentages of omnivory, arboreal quadrupedalism, terrestrial/arboreal quadrupedalism and two body mass classes accounted for 98.1% of the variation in ecological structure. Transitional communities were projected in intermediate positions by the discriminant model. Main conclusions Our analysis demonstrates that the broad‐scale habitat and climate variables captured by the ecoregion model have shaped the assembly of non‐volant mammal communities in Madagascar over evolutionary time. The spatial pattern is consistent with ecological sorting of species ranges along environmental gradients. Historical processes, such as recent extinction and migration, may have also affected the structure of mammal communities, although these factors have played a secondary role.