|Home | About | Journals | Submit | Contact Us | Français|
The European Nucleotide Archive (ENA; http://www.ebi.ac.uk/ena), Europe's primary nucleotide sequence resource, captures and presents globally comprehensive nucleic acid sequence and associated information. Covering the spectrum from raw data to assembled and functionally annotated genomes, the ENA has witnessed a dramatic growth resulting from advances in sequencing technology and ever broadening application of the methodology. During 2011, we have continued to operate and extend the broad range of ENA services. In particular, we have released major new functionality in our interactive web submission system, Webin, through developments in template-based submissions for annotated sequences and support for raw next-generation sequence read submissions.
The European Nucleotide Archive (ENA) is maintained and developed at the European Molecular Biology Laboratory's European Bioinformatics Institute (EMBL-EBI) and serves as Europe's primary repository for nucleotide sequence and associated information. Content spans raw sequence reads from all sequencing platforms, read alignments, assembly information and submitted functional annotation. Providing both the permanent scientific record as a complement to literature publication process and a forum for early sharing of pre-publication data, the ENA serves as a critical foundation for the global bioinformatics data infrastructure. Globally comprehensive coverage is assured through long-standing data exchange agreements with the DNA Databank of Japan (DDBJ) (1) and the United States National Institutes of Health National Center for Biotechnology Information (NCBI) (2) under the International Nucleotide Sequence Database Collaboration (3; http://www.insdc.org/).
Underlying ENA are a number of core databases, including the Sequence Read Archive for raw reads and read alignments from next generation sequencing platforms (4) and EMBL-Bank for high level assembly information, assembled sequences and functional annotation. ENA services are numerous: we provide submission tools, both the web-based Webin system and programmatic interfaces; we offer search technologies, such as the newly developed rapid ENA sequence similarity search (http://www.ebi.ac.uk/ena/search) and text-based search tools (http://www.ebi.ac.uk/ena); we present integrated access to all ENA content through the ENA Browser, which offers both web browsing and REST access (http://www.ebi.ac.uk/ena/about/browser). We are highly responsive in the development of new technologies and services to adapt to changes in sequencing technology and user requirements: we are leading a community-facing sequence read compression initiative, CRAM (5; http://www.ebi.ac.uk/ena/about/cram_toolkit); we are developing anencrypted BAM read alignment server that supports reference coordinate-based lookups of controlled acess reads by region; we are active in the development of data warehousing methodologies to provide real-time access to the massive data sets that we store (e.g. the ENA Taxon Portal; http://www.ebi.ac.uk/ena/data/view/Taxon:Eukaryota).
In this article, we comment on content and report briefly on means by which ENA data can be accessed. We then focus on major developments in our Webin submission system in the areas of template-based submissions of annotated and assembled sequences and raw next generation sequence read submission. We also announce the introduction of a sequence length limit for submission of assembled sequences.
At the time of going to press, ENA contains 346598699035nt of assembled sequence in 220504007 assembled sequence entries (See EMBL-Bank release notes at http://www.ebi.ac.uk/embl/Documentation/Release_notes/current/relnotes.html) and more than 100 terabases of raw next generation sequence reads (Figure 1A and B).
Notable datasets submitted to ENA during 2011 include assemblies of Gorilla gorilla (FR853080-FR853106), atlantic cod, Gadus morhua (Project:41391), Vine, Vitis vinifera (Project:18785), Takifugu rubripes (Project:1434), Macaca fascicularis (FR874244-FR874264), medieval mitochondria and Yersinia plasmids (6; HE576978-HE576987), raw genomic reads from 18 lines of Arabidopsis thaliana (7; ERP000565), Staphylococcus aureus (8; ERP000528) and Mus musculus ES cells (9; ERP000570) and transcriptomicreads from multiple Silene species (10; ERP000371).
Full ENA content is made available through an integrated platform, the ENA Browser, that supports discovery (text search, sequence similarity search, taxon lookup, etc.) and retrieval of records interactively (through web browsing and programatically under RESTful URLs). Full details are available from http://www.ebi.ac.uk/ena/about/browser. Records are made available in a selection of appropriate formats that include EMBL-Bank flat file, fasta and XML for assembled and annotated sequences, Fastq for sequence reads and Darwin Core for taxon records (http://www.ebi.ac.uk/ena/about/formats). In addition, we support both ftp and Aspera protocols for network transfers of large raw data sets (ftp://ftp.sra.ebi.ac.uk) and offer a variety of data products over ftp for other areas of ENA content (ftp://ftp.ebi.ac.uk/pub/databases/embl and ftp://ftp.ebi.ac.uk/pub/databases/ena)
Apre-tailored template system was introduced in our Webin submission framework in 2009 for annotated sequence submissions and has been expanded during 2011 with the release of nine new templates. These templates have been designed for the most frequent types of sequence submissions and reached 15 in number in September 2011. When using the templates, submitters provide nucleotide sequences with associated annotation through spread sheets or Fastq files with pre-defined mandatory and optional fields, a process that significantly reduces the overall complexity of the submissions process for both the submitter and the ENA curator. Some advantages of the new system include the ability to choose from a small number of variables, functionalities that prevent the need for repetitive entry of information constant across all records in a data set and straightforward validation before data submission. The template concept has shown growing popularity since its launch versus the traditional system (which remains available for a limited time). Under the traditional system, submitters were able to annotate their entries with the full INSDC-approved features and qualifiers either one entry at a time or by defining with an ENA curator a specific template for each submission. This was useful for annotating small submissions in great detail but did not cater efficiently for larger-scale submissions of same-type data. Figure 2 shows the usage of the available submission systems between 2009 and 2011 and Table 1 shows the currently available templates.
As part of these developments, ENA is also facilitating the submission of marker gene sequences compliant with a community standard that has been developed by the Genomic Standards Consortium (GSC), called the Minimal Information about a MARKer gene Sequence Standard (MIMARKS) (11, 12). MIMARKS provides a minimal set of required information fields essential for downstream reuse of the data. The last two templates in Table 1 have been designed for submissions of MIMARKS-compliant data.
Further improvements to the submissions system for annotated sequences will continue in 2012 and beyond.
To complement the existing programmatic SRA REST submission interface, we have recently extended the Webin system to support submissions of raw next generation sequencing reads to the SRA. Unlike the SRA REST interface, which is targeted for large-scale sequence submitters and allows direct programmatic interaction between external LIMS systems and the SRA database at EBI, this new component of Webin is designed for interactive use. Users work through a web interface to create studies, samples and experiments, to update submitted metadata and to release previously submitted data to the public. Importantly, all metadata are submitted either by uploading or editing spreadsheets. While SRA REST submitters are fully exposed to the underlying SRA XML-data model, the SRA submission functionality in Webin completely hides this complexity. For example, during a raw sequence submission process, users are asked to define their raw data file format and are then presented with a spreadsheet, which can be either uploaded or filled with the required additional information (Figure 3).
The SRA submission component of Webin is under active development and new improvements are deployed weekly. Forthcoming improvements include support for European Genome–Phenome Archive submissions for controlled access raw sequence data, support for checklist for provision of community standard compliant meta data and numerous usability additions.
ENA will introduce a sequence length limit for submissions of assembled sequences. From January 2012, ENA will accept sequences <100bp only if they fall into one of the following sequence categories of ‘Ancient DNA’, ‘non-coding-RNA’, ‘Microsatellites’ or ‘Complete Exons’. Exceptions require the submitter to demonstrate that a peer-reviewed journal has accepted a manuscript by the submitter, confirming the relevance of the short sequences to the scientific community. A validation step will be implemented in Webin to facilitate implementation of this requirement. We encourage submitters to check our website for further forthcoming changes announcements (http://www.ebi.ac.uk/ena/about/forthcoming_changes)
The ENA team provides advice and guidance regarding ENA services by email through firstname.lastname@example.org. Feedback and suggestions related to all of our services are very welcome at the same email address. We also operate a variety of hands-on training programmes, for which details are available at http://www.ebi.ac.uk/training. We strongly encourage submitters to take our survey (http://www.surveymonkey.com/s/ENA_User_Survey_2011) and help us to improve our service.
European Molecular Biology Laboratory; FP7 Programme of the European Commission; WellcomeTrust and Biotechnology and Biological Sciences Research Council (BBSRC). Funding for open access charge: European Molecular Biology Laboratory.
Conflict of interest statement. None declared.