Bravo à Chrystelle Delors et Valentine Rech de Laval qui ont remporté les prix poster décerné par la SFBI !
Le poster de Chrystelle :
In a perspective of conservation, considering the state of genetic resources of species is often of critical importance. To this end, a main issue consists in the ability to cost-efficiently genotype a high number of molecular markers in a high number of individuals, so one reinforces the pertinence of subsequent biological interpretations. The prior challenge is the access to a catalog or relevant and suitable molecular markers, which could be particularly complex when considering non-model species for which genetic knowledge is scarce. One of our objectives is to develop a set of specific markers for several neotropical fish populations (Maroni river, French Guiana) whose dispersal abilities and population structuration are almost entirely unknown. Indeed those species are often of high cultural and alimentary value for autochtonous people, for instance Wayana amerindians who are hoping for a sustainable management of those artisanal fisheries.
Here, we present our strategy, that integrates RAD-seq of pooled individuals for SNP calling, and should make us able to cost-efficiently develop and genotype a reasonable number of SNPs markers for each species of interest, sufficient to adress population genetics issues. We mainly focused on some of the relevant questions one needs to think about before establishing the computational process (pipeline) that will be adapted to extract those markers from NGS data. We also present some of the most widely used high-throughput genotyping methods, as well as their pros and cons.
Le parcours de Chrystelle :
J'ai effectué une licence de Biologie des Organismes et des Populations à l'Université Claude Bernard Lyon 1, suite à quoi je me suis spécialisée dans les systèmes aquatiques et marins. Je suis actuellement en thèse au sein de l'UMR Ecologie et Santé des Ecosystèmes (INRA/Agrocampus Ouest, Rennes) et mon travail porte sur l'étude de la structuration génétique et du potentiel de dispersion chez plusieurs espèces de poissons néotropicaux, dans une perspective de conservation.
Le poster de Valentine :
neXtProt (http://www.nextprot.org) is a comprehensive human proteincentric knowledgebase. Its webbased platform offers to its users a seamless integration of and navigation through proteinrelated data. Focused solely on human proteins, neXtProt aims to provide a state of the art resource for the representation of human biology by capturing a wide range of data, precise annotations, fully traceable data provenance and a web interface, which enables researchers to find and view information in a comprehensive manner.
neXtProt is both a new and an old resource: new, because we try to create an innovative integrative resource around human proteins and old, because we are building it on top of the highquality solid work that has been the hallmark of UniProtKB/SwissProt. The extensive efforts made by SwissProt to functionally annotate human proteins and curate their sequences and many other features is the foundation on which neXtProt relies. However, this is not enough to populate a resource that needs to address the complexity of the universe of human proteins. In order to remedy this, neXtProt continuously adds new content to the database. The major data sources include Bgee, Human Protein Atlas (HPA), Peptide Atlas, SRMAtlas, UniProtKB, GOA, COSMIC, and IntAct.
neXtProt allows querying data to access entries, publications and terminology. There are several ways to query neXtProt content.
The simple search system (http://search.nextprot.org) is a Googlelike full text search which allows you to search the data in neXtProt using Solr technology. The so called advanced search system uses the SPARQL Protocol and RDF Query Language (SPARQL) to access all the neXtProt protein entry data. This search was designed to support the retrieval of proteins based on highly precise criteria taking into account the richness of the annotations and evidences. Tools were developed and integrated in our user interface to help users to learn how and work out SPARQL queries (http://snorql.nextprot.org, http://sparqlplayground.nextprot.org).
It is also possible querying data via a REST API (http://api.nextprot.org). This decouples the database from all our services; in particular, the search and the export services. The API services make it possible for our users to develop their own applications / tools that will benefit from neXtProt's clean and highquality data.
neXtProt data are downloadable in two ways. After querying data with simple or advanced search, you can download one or all entries of the displayed result of the search. Entries can be exported in their entirety, or the users can customize which content they wish to export, for instance the sequence or a subset of annotation types like PTMs or expression profiles. Several formats are available: FASTA for sequences, and XLS, JSON and XML for entries. The second possibility is to download current and past releases of neXtProt, controlled vocabularies and ontologies developed specifically for neXtProt, as well as other files from the FTP site (ftp://ftp.nextprot.org) in different formats (FASTA, PEFF, XML, RDF, etc.).
It is possible to explore data via our web interface. We are developing viewers for the display of our data as modular generic components to make them reusable by the life sciences community independently of neXtProt. One of them is a viewer displaying detailed information about the mass spectrometryderived peptides observed in a protein sequence. Another one displays the different features of a specific protein. These new tools are pretty generic and could be applied to many other contexts. The source code of our components is available on Github (http://github.com/caliphosib).
The data in neXtProt can be analyzed using tools: BLAST and a list manager. We provides access to a simple BLAST implementation to find other entries in neXtProt containing the same or a similar user sequence. It is possible to create a collection of protein entries. neXtProt allows you to create and manage private lists.
The big challenge is to be able to understand how a given set of proteins share or differ in their various features. Indeed data is useful, but we want neXtProt to be much more than a wellorganized comprehensive data repository. In addition to enhanced search capabilities, we offer tools that help to make sense of the contents.
Le parcours de Valentine :
Après un master en "Génomique Structurale et Bioinformatique" à l'université de Strasbourg fini en 2009, j'ai fait une thèse à l'IBCP-BMSSI de l'université Lyon. Intitulée "Analyse bioinformatique des protéines BCL-2 et développement de la base de connaissance dédiée, BCL2DB", je l'ai soutenue en 2013. Depuis 2014, je travaille à l'Université de Lausanne (SIB-UNIL) sur le projet Bgee (dataBase for Gene Expression Evolution) en tant que développeur bioinformatique java. Et depuis 2015, je travaille aussi, en tant que post-doctorante, à l'Université de Genève (SIB-UNIGE), sur le projet neXtProt (une base de données consacrée aux les protéines humaines).