AutoFACT: An Automatic Functional Annotation and Classification Tool

Affiliation: Centre Robert-Cedergren de l'Université de Montréal en bio-informatique et génomique & Département de biochimie, Université de Montréal

Annotation XML et interrogation de corpus pour l’étude de la conceptualisation métaphorique

Département de linguistique et de traduction

Annotation des ARN non codants du génome de Candida albicans par méthode bioinformatique

La bio-informatique est un champ pluridisciplinaire qui utilise la biologie, l’informatique, la physique et les mathématiques pour résoudre des problèmes posés par la biologie. L’une des thématiques de la bio-informatique est l’analyse des séquences génomiques et la prédiction de gènes d’ARN non codants. Les ARN non codants sont des molécules d’ARN qui sont transcrites mais pas traduites en protéine et qui ont une fonction dans la cellule. Trouver des gènes d’ARN non codants par des techniques de biochimie et de biologie moléculaire est assez difficile et relativement coûteux. Ainsi, la prédiction des gènes d’ARNnc par des méthodes bio-informatiques est un enjeu important. Cette recherche décrit un travail d’analyse informatique pour chercher des nouveaux ARNnc chez le pathogène Candida albicans et d’une validation expérimentale. Nous avons utilisé comme stratégie une analyse informatique combinant plusieurs logiciels d’identification d’ARNnc. Nous avons validé un sous-ensemble des prédictions informatiques avec une expérience de puces à ADN couvrant 1979 régions du génome. Grace à cette expérience nous avons identifié 62 nouveaux transcrits chez Candida albicans. Ce travail aussi permit le développement d’une méthode d’analyse pour des puces à ADN de type tiling array. Ce travail présente également une tentation d’améliorer de la prédiction d’ARNnc avec une méthode se basant sur la recherche de motifs d’ARN dans les séquences.

Annotation syntaxico-sémantique des actants en corpus spécialisé

L’annotation en rôles sémantiques est une tâche qui permet d’attribuer des étiquettes de rôles telles que Agent, Patient, Instrument, Lieu, Destination etc. aux différents participants actants ou circonstants (arguments ou adjoints) d’une lexie prédicative. Cette tâche nécessite des ressources lexicales riches ou des corpus importants contenant des phrases annotées manuellement par des linguistes sur lesquels peuvent s’appuyer certaines approches d’automatisation (statistiques ou apprentissage machine). Les travaux antérieurs dans ce domaine ont porté essentiellement sur la langue anglaise qui dispose de ressources riches, telles que PropBank, VerbNet et FrameNet, qui ont servi à alimenter les systèmes d’annotation automatisés. L’annotation dans d’autres langues, pour lesquelles on ne dispose pas d’un corpus annoté manuellement, repose souvent sur le FrameNet anglais. Une ressource telle que FrameNet de l’anglais est plus que nécessaire pour les systèmes d’annotation automatisé et l’annotation manuelle de milliers de phrases par des linguistes est une tâche fastidieuse et exigeante en temps. Nous avons proposé dans cette thèse un système automatique pour aider les linguistes dans cette tâche qui pourraient alors se limiter à la validation des annotations proposées par le système. Dans notre travail, nous ne considérons que les verbes qui sont plus susceptibles que les noms d’être accompagnés par des actants réalisés dans les phrases. Ces verbes concernent les termes de spécialité d’informatique et d’Internet (ex. accéder, configurer, naviguer, télécharger) dont la structure actancielle est enrichie manuellement par des rôles sémantiques. La structure actancielle des lexies verbales est décrite selon les principes de la Lexicologie Explicative et Combinatoire, LEC de Mel’čuk et fait appel partiellement (en ce qui concerne les rôles sémantiques) à la notion de Frame Element tel que décrit dans la théorie Frame Semantics (FS) de Fillmore. Ces deux théories ont ceci de commun qu’elles mènent toutes les deux à la construction de dictionnaires différents de ceux issus des approches traditionnelles. Les lexies verbales d’informatique et d’Internet qui ont été annotées manuellement dans plusieurs contextes constituent notre corpus spécialisé. Notre système qui attribue automatiquement des rôles sémantiques aux actants est basé sur des règles ou classificateurs entraînés sur plus de 2300 contextes. Nous sommes limités à une liste de rôles restreinte car certains rôles dans notre corpus n’ont pas assez d’exemples annotés manuellement. Dans notre système, nous n’avons traité que les rôles Patient, Agent et Destination dont le nombre d’exemple est supérieur à 300. Nous avons crée une classe que nous avons nommé Autre où nous avons rassemblé les autres rôles dont le nombre d’exemples annotés est inférieur à 100. Nous avons subdivisé la tâche d’annotation en sous-tâches : identifier les participants actants et circonstants et attribuer des rôles sémantiques uniquement aux actants qui contribuent au sens de la lexie verbale. Nous avons soumis les phrases de notre corpus à l’analyseur syntaxique Syntex afin d’extraire les informations syntaxiques qui décrivent les différents participants d’une lexie verbale dans une phrase. Ces informations ont servi de traits (features) dans notre modèle d’apprentissage. Nous avons proposé deux techniques pour l’identification des participants : une technique à base de règles où nous avons extrait une trentaine de règles et une autre technique basée sur l’apprentissage machine. Ces mêmes techniques ont été utilisées pour la tâche de distinguer les actants des circonstants. Nous avons proposé pour la tâche d’attribuer des rôles sémantiques aux actants, une méthode de partitionnement (clustering) semi supervisé des instances que nous avons comparée à la méthode de classification de rôles sémantiques. Nous avons utilisé CHAMÉLÉON, un algorithme hiérarchique ascendant.

Automatic Image Annotation Using SURF Descriptors

In recent years there is an apparent shift in research from content based image retrieval (CBIR) to automatic image annotation in order to bridge the gap between low level features and high level semantics of images. Automatic Image Annotation (AIA) techniques facilitate extraction of high level semantic concepts from images by machine learning techniques. Many AIA techniques use feature analysis as the first step to identify the objects in the image. However, the high dimensional image features make the performance of the system worse. This paper describes and evaluates an automatic image annotation framework which uses SURF descriptors to select right number of features and right features for annotation. The proposed framework uses a hybrid approach in which k-means clustering is used in the training phase and fuzzy K-NN classification in the annotation phase. The performance of the system is evaluated using standard metrics.

Capturing Emergent Semantics from Social Annotation Systems

The ongoing growth of the World Wide Web, catalyzed by the increasing possibility of ubiquitous access via a variety of devices, continues to strengthen its role as our prevalent information and commmunication medium. However, although tools like search engines facilitate retrieval, the task of finally making sense of Web content is still often left to human interpretation. The vision of supporting both humans and machines in such knowledge-based activities led to the development of different systems which allow to structure Web resources by metadata annotations. Interestingly, two major approaches which gained a considerable amount of attention are addressing the problem from nearly opposite directions: On the one hand, the idea of the Semantic Web suggests to formalize the knowledge within a particular domain by means of the "top-down" approach of defining ontologies. On the other hand, Social Annotation Systems as part of the so-called Web 2.0 movement implement a "bottom-up" style of categorization using arbitrary keywords. Experience as well as research in the characteristics of both systems has shown that their strengths and weaknesses seem to be inverse: While Social Annotation suffers from problems like, e. g., ambiguity or lack or precision, ontologies were especially designed to eliminate those. On the contrary, the latter suffer from a knowledge acquisition bottleneck, which is successfully overcome by the large user populations of Social Annotation Systems. Instead of being regarded as competing paradigms, the obvious potential synergies from a combination of both motivated approaches to "bridge the gap" between them. These were fostered by the evidence of emergent semantics, i. e., the self-organized evolution of implicit conceptual structures, within Social Annotation data. While several techniques to exploit the emergent patterns were proposed, a systematic analysis - especially regarding paradigms from the field of ontology learning - is still largely missing. This also includes a deeper understanding of the circumstances which affect the evolution processes. This work aims to address this gap by providing an in-depth study of methods and influencing factors to capture emergent semantics from Social Annotation Systems. We focus hereby on the acquisition of lexical semantics from the underlying networks of keywords, users and resources. Structured along different ontology learning tasks, we use a methodology of semantic grounding to characterize and evaluate the semantic relations captured by different methods. In all cases, our studies are based on datasets from several Social Annotation Systems. Specifically, we first analyze semantic relatedness among keywords, and identify measures which detect different notions of relatedness. These constitute the input of concept learning algorithms, which focus then on the discovery of synonymous and ambiguous keywords. Hereby, we assess the usefulness of various clustering techniques. As a prerequisite to induce hierarchical relationships, our next step is to study measures which quantify the level of generality of a particular keyword. We find that comparatively simple measures can approximate the generality information encoded in reference taxonomies. These insights are used to inform the final task, namely the creation of concept hierarchies. For this purpose, generality-based algorithms exhibit advantages compared to clustering approaches. In order to complement the identification of suitable methods to capture semantic structures, we analyze as a next step several factors which influence their emergence. Empirical evidence is provided that the amount of available data plays a crucial role for determining keyword meanings. From a different perspective, we examine pragmatic aspects by considering different annotation patterns among users. Based on a broad distinction between "categorizers" and "describers", we find that the latter produce more accurate results. This suggests a causal link between pragmatic and semantic aspects of keyword annotation. As a special kind of usage pattern, we then have a look at system abuse and spam. While observing a mixed picture, we suggest that an individual decision should be taken instead of disregarding spammers as a matter of principle. Finally, we discuss a set of applications which operationalize the results of our studies for enhancing both Social Annotation and semantic systems. These comprise on the one hand tools which foster the emergence of semantics, and on the one hand applications which exploit the socially induced relations to improve, e. g., searching, browsing, or user profiling facilities. In summary, the contributions of this work highlight viable methods and crucial aspects for designing enhanced knowledge-based services of a Social Semantic Web.

Annotation and characterization of the Plasmodium vivax rhoptry neck protein 4 (PvRON4)

Background: The tight junction (TJ) is one of the most important structures established during merozoite invasion of host cells and a large amount of proteins stored in Toxoplasma and Plasmodium parasites’ apical organelles are involved in forming the TJ. Plasmodium falciparum and Toxoplasma gondii apical membrane antigen 1 (AMA-1) and rhoptry neck proteins (RONs) are the two main TJ components. It has been shown that RON4 plays an essential role during merozoite and sporozoite invasion to target cells. This study has focused on characterizing a novel Plasmodium vivax rhoptry protein, RON4, which is homologous to PfRON4 and PkRON4. Methods: The ron4 gene was re-annotated in the P. vivax genome using various bioinformatics tools and taking PfRON4 and PkRON4 amino acid sequences as templates. Gene synteny, as well as identity and similarity values between open reading frames (ORFs) belonging to the three species were assessed. The gene transcription of pvron4, and the expression and localization of the encoded protein were also determined in the VCG-1 strain by molecular and immunological studies. Nucleotide and amino acid sequences obtained for pvron4 in VCG-1 were compared to those from strains coming from different geographical areas. Results: PvRON4 is a 733 amino acid long protein, which is encoded by three exons, having similar transcription and translation patterns to those reported for its homologue, PfRON4. Sequencing PvRON4 from the VCG-1 strain and comparing it to P. vivax strains from different geographical locations has shown two conserved regions separated by a low complexity variable region, possibly acting as a “smokescreen”. PvRON4 contains a predicted signal sequence, a coiled-coil α-helical motif, two tandem repeats and six conserved cysteines towards the carboxyterminus and is a soluble protein lacking predicted transmembranal domains or a GPI anchor. Indirect immunofluorescence assays have shown that PvRON4 is expressed at the apical end of schizonts and co-localizes at the rhoptry neck with PvRON2.

An extensible automated protein annotation tool: standardizing input and output using validated XML

Motivation: There is a frequent need to apply a large range of local or remote prediction and annotation tools to one or more sequences. We have created a tool able to dispatch one or more sequences to assorted services by defining a consistent XML format for data and annotations. Results: By analyzing annotation tools, we have determined that annotations can be described using one or more of the six forms of data: numeric or textual annotation of residues, domains (residue ranges) or whole sequences. With this in mind, XML DTDs have been designed to store the input and output of any server. Plug-in wrappers to a number of services have been written which are called from a master script. The resulting APATML is then formatted for display in HTML. Alternatively further tools may be written to perform post-analysis.

Semi-automatic annotation and retrieval of visual content using the topic map technology

There are still major challenges in the area of automatic indexing and retrieval of multimedia content data for very large multimedia content corpora. Current indexing and retrieval applications still use keywords to index multimedia content and those keywords usually do not provide any knowledge about the semantic content of the data. With the increasing amount of multimedia content, it is inefficient to continue with this approach. In this paper, we describe the project DREAM, which addresses such challenges by proposing a new framework for semi-automatic annotation and retrieval of multimedia based on the semantic content. The framework uses the Topic Map Technology, as a tool to model the knowledge automatically extracted from the multimedia content using an Automatic Labelling Engine. We describe how we acquire knowledge from the content and represent this knowledge using the support of NLP to automatically generate Topic Maps. The framework is described in the context of film post-production.

Proteogenomics and in silico structural and functional annotation of the barley powdery mildew Blumeria graminis f. sp. hordei

Blumeria graminis is an economically important obligate plant-pathogenic fungus, whose entire genome was recently sequenced and manually annotated using ab initio in silico predictions [7]. Employing large scale proteogenomic analysis we are now able to verify independently the existence of proteins predicted by 24% of open reading frame models. We compared the haustoria and sporulating hyphae proteomes and identified 71 proteins exclusively in haustoria, the feeding and effector-delivery organs of the pathogen. These proteins are ‘significantly smaller than the rest of the protein pool and predicted to be secreted. Most do not share any similarities with Swiss–Prot or Trembl entries nor possess any identifiable Pfam domains. We used a novel automated prediction pipeline to model the 3D structures of the proteins, identify putative ligand binding sites and predict regions of intrinsic disorder. This revealed that the protein set found exclusively in haustoria is significantly less disordered than the rest of the identified Blumeria proteins or random (and representative) protein sets generated from the yeast proteome. For most of the haustorial proteins with unknown functions no good templates could be found, from which to generate high quality models. Thus, these unknown proteins present potentially new protein folds that can be specific to the interaction of the pathogen with its host.

Predicting protein structures and structural annotation of proteomes

Protein structure prediction methods aim to predict the structures of proteins from their amino acid sequences, utilizing various computational algorithms. Structural genome annotation is the process of attaching biological information to every protein encoded within a genome via the production of three-dimensional protein models.