Steffen Möller

A novel method for automatic functional annotation of proteins.

MOTIVATION To cope with the increasing amount of sequence data, reliable automatic annotation tools are required. The TrEMBL database contains together with SWISS-PROT nearly all publicly available protein sequences, but in contrast to SWISS-PROT only limited functional annotation. To improve this situation, we had to develop a method of automatic annotation that produces highly reliable functional prediction using the language and the syntax of SWISS-PROT. RESULTS An algorithm was developed and successfully used for the automatic annotation of a testset of unknown proteins. The predicted information included description, function, catalytic activity, cofactors, pathway, subcellular location, quaternary structure, similarity to other protein, active sites, and keywords. The algorithm showed a low coverage (10%), but a high specificity and reliability. AVAILABILITY The results can be obtained by anonymous ftp from ftp.ebi.ac.uk/pub/databases/sp_tr_nrdb. The source code is available on request from the authors.

Prediction of Apoptosis Protein Locations with Genetic Algorithms and Support Vector Machines Through a New Mode of Pseudo Amino Acid Composition

Apoptosis is an essential process for controlling tissue homeostasis by regulating a physiological balance between cell proliferation and cell death. The subcellular locations of proteins performing the cell death are determined by mostly independent cellular mechanisms. The regular bioinformatics tools to predict the subcellular locations of such apoptotic proteins do often fail. This work proposes a model for the sorting of proteins that are involved in apoptosis, allowing us to both the prediction of their subcellular locations as well as the molecular properties that contributed to it. We report a novel hybrid Genetic Algorithm (GA)/Support Vector Machine (SVM) approach to predict apoptotic protein sequences using 119 sequence derived properties like frequency of amino acid groups, secondary structure, and physicochemical properties. GA is used for selecting a near-optimal subset of informative features that is most relevant for the classification. Jackknife cross-validation is applied to test the predictive capability of the proposed method on 317 apoptosis proteins. Our method achieved 85.80% accuracy using all 119 features and 89.91% accuracy for 25 features selected by GA. Our models were examined by a test dataset of 98 apoptosis proteins and obtained an overall accuracy of 90.34%. The results show that the proposed approach is promising; it is able to select small subsets of features and still improves the classification accuracy. Our model can contribute to the understanding of programmed cell death and drug discovery. The software and dataset are available at http://www.inb.uni-luebeck.de/tools-demos/apoptosis/GASVM.

Genetic identification and functional validation of FcγRIV as key molecule in autoantibody-induced tissue injury†

Autoantibody‐mediated diseases are clinically heterogeneous and often fail conventional therapeutic strategies. Gene expression profiling has helped to identify new molecular pathways in these diseases, although their potential as treatment targets largely remains to be functionally validated. Based on weighted gene co‐expression network analysis, we determined the transcriptional network in experimental epidermolysis bullosa acquisita (EBA), a paradigm of an antibody‐mediated organ‐specific autoimmune disease characterized by autoantibodies directed against type VII collagen. We identified 33 distinct and differentially expressed modules, including Fcγ receptor (FcγR) IV and components of the neutrophil‐associated enzyme system in autoantibody transfer‐induced EBA. Validation experiments, including functional analysis, demonstrated that FcγRIV expression on neutrophils crucially contributes to autoantibody‐induced tissue injury in the transfer model of EBA. Mice lacking the common γ‐chain of activating FcγRs, deficient in FcγRIV or treated with FcγRIV function blocking antibody, but not mice deficient in FcγRI, FcγRIIB, FcγRIII or both FcγRI and FcγRIII, were effectively protected from EBA. Skin disease was restored in γ‐chain‐deficient mice locally reconstituted with neutrophils from wild‐type, but not from γ‐chain‐deficient, mice. Our findings both genetically and functionally identify a novel disease‐related molecule, FcγRIV, in an autoantibody‐mediated disorder, which may be of importance for the development of novel targeted therapies. Copyright

Cytoskeletal Rearrangements in Synovial Fibroblasts as a Novel Pathophysiological Determinant of Modeled Rheumatoid Arthritis

Rheumatoid arthritis is a chronic inflammatory disease with a high prevalence and substantial socioeconomic burden. Despite intense research efforts, its aetiology and pathogenesis remain poorly understood. To identify novel genes and/or cellular pathways involved in the pathogenesis of the disease, we utilized a well-recognized tumour necrosis factor-driven animal model of this disease and performed high-throughput expression profiling with subtractive cDNA libraries and oligonucleotide microarray hybridizations, coupled with independent statistical analysis. This twin approach was validated by a number of different methods in other animal models of arthritis as well as in human patient samples, thus creating a unique list of disease modifiers of potential therapeutic value. Importantly, and through the integration of genetic linkage analysis and Gene Ontology–assisted functional discovery, we identified the gelsolin-driven synovial fibroblast cytoskeletal rearrangements as a novel pathophysiological determinant of the disease.

EDITtoTrEMBL: a distributed approach to high-quality automated protein sequence annotation.

SUMMARY Many databases in molecular biology face the problem that the ever increasing rate of data production can no longer be handled by traditional methods, especially human curation. Therefore, a number of projects are currently investigating methods for automated sequence annotation. This paper describes the EBIs approach to this problem for protein sequences by integration of arbitrary analysis programs into a distributed and highly flexible environment. Our software framework allows an individual treatment of sequences depending on their particular properties, which is achieved through a high-level description of the preconditions and capabilities of analysing modules. This not only improves the overall performance of the annotation process, as unnecessary steps are avoided, but also enhances its quality since dependencies between different modules are taken into account. We have implemented a prototype and use it in the production of TrEMBL releases. AVAILABILITY Upon request.

Genome-wide mapping of gene–microbiota interactions in susceptibility to autoimmune skin blistering

Susceptibility to chronic inflammatory diseases is determined by immunogenetic and environmental risk factors. Resident microbial communities often differ between healthy and diseased states, but whether these differences are of primary aetiological importance or secondary to the altered inflammatory environment remains largely unknown. Here we provide evidence for host gene–microbiota interactions contributing to disease risk in a mouse model of epidermolysis bullosa acquisita, an autoantibody-induced inflammatory skin disease. Using an advanced intercross, we identify genetic loci contributing to skin microbiota variability, susceptibility to skin blistering and their overlap. Furthermore, by treating bacterial species abundances as covariates with disease we reveal a novel disease locus. The majority of the identified covariate taxa are characterized by reduced abundance being associated with increased disease risk, providing evidence of a primary role in protection from disease. Further characterization of these putative probiotic species or species assemblages offers promising potential for preventative and therapeutic treatment development.

Community-driven computational biology with Debian Linux

BackgroundThe Open Source movement and its technologies are popular in the bioinformatics community because they provide freely available tools and resources for research. In order to feed the steady demand for updates on software and associated data, a service infrastructure is required for sharing and providing these tools to heterogeneous computing environments.ResultsThe Debian Med initiative provides ready and coherent software packages for medical informatics and bioinformatics. These packages can be used together in Taverna workflows via the UseCase plugin to manage execution on local or remote machines. If such packages are available in cloud computing environments, the underlying hardware and the analysis pipelines can be shared along with the software.ConclusionsDebian Med closes the gap between developers and users. It provides a simple method for offering new releases of software and data resources, thus provisioning a local infrastructure for computational biology. For geographically distributed teams it can ensure they are working on the same versions of tools, in the same conditions. This contributes to the world-wide networking of researchers.

Tools and data services registry: a community effort to document bioinformatics resources

Life sciences are yielding huge data sets that underpin scientific discoveries fundamental to improvement in human health, agriculture and the environment. In support of these discoveries, a plethora of databases and tools are deployed, in technically complex and diverse implementations, across a spectrum of scientific disciplines. The corpus of documentation of these resources is fragmented across the Web, with much redundancy, and has lacked a common standard of information. The outcome is that scientists must often struggle to find, understand, compare and use the best resources for the task at hand. Here we present a community-driven curation effort, supported by ELIXIR—the European infrastructure for biological information—that aspires to a comprehensive and consistent registry of information about bioinformatics resources. The sustainable upkeep of this Tools and Data Services Registry is assured by a curation effort driven by and tailored to local needs, and shared amongst a network of engaged partners. As of November 2015, the registry includes 1785 resources, with depositions from 126 individual registrations including 52 institutional providers and 74 individuals. With community support, the registry can become a standard for dissemination of information about bioinformatics resources: we welcome everyone to join us in this common endeavour. The registry is freely available at https://bio.tools.

Fine Mapping of Collagen-Induced Arthritis Quantitative Trait Loci in an Advanced Intercross Line

The generation of advanced intercross lines (AIL) is a powerful approach for high-resolution fine mapping of quantitative trait loci (QTLs), because they accumulate much more recombination events compared with conventional F2 intercross and N2 backcross. However, the application of this approach is severely hampered by the requirements of excessive resources to maintain such crosses, i.e., in terms of animal care, space, and time. Therefore, in this study, we produced an AIL to fine map collagen-induced arthritis (CIA) QTLs using comparatively limited resources. We used only 308 (DBA/1 × FVB/N)F11/12 AIL mice to refine QTLs controlling the severity and onset of arthritis as well as the Ab response and T cell subset in CIA, namely Cia2, Cia27, and Trmq3. These QTLs were originally identified in (DBA/1 × FVB/N)F2 progeny. The confidence intervals of the three QTLs were refined from 40, 43, and 48 Mb to 12, 4.1, and 12 Mb, respectively. The data were complemented by the use of another QTL fine-mapping approach, haplotype analysis, to further refine Cia2 into a 2-Mb genomic region. To aid in the search for candidate genes for the QTLs, genome-wide expression profiling was performed to identify strain-specific differentially expressed genes within the confidence intervals. Of the 1396 strain-specific differentially expressed genes, 3, 3, and 12 genes were within the support intervals of the Cia2, Cia27, and Trmq3, respectively. In addition, this study revealed that Cia27 and Trmq3 controlling anti-CII IgG2a Ab and CD4:CD8 T cell ratio, respectively, also regulated CIA clinical phenotypes.

Expressionview: visualization of quantitative trait loci and gene-expression data in Ensembl

We present here a software tool for combined visualization of gene-expression data and quantitative trait loci (QTL). The application is implemented as an extension to the Ensembl project and caters for a direct transition from microarray experiments of gene or protein expression levels to the genomic context of individual genes and QTL. It supports the visualization of gene clusters and the selection of functional candidate genes in the context of research on complex traits.