Leiden University Medical Center
P.O. Box 9600, 2300RC Leiden Visiting address: Building 2, Room R04-005, Einthovenweg 20, Leiden.
Dr. P.E.M. Taschner
0031 (0)71 5269424
Our department has performed pioneering work on the topic of gene variant databases by establishing the Leiden Muscular Dystrophy pages (LMDp, http://www.DMD.nl). Besides providing general background information on our research and diagnostics on Duchenne/Becker muscular dystrophy, we set up one of the first internet-based gene variant databases (also known as LSDBs, Locus Specific DataBases). The gene variant database group is part of our departmental Bioinformatics Research Group and developed: 1) the freely available open source generic Leiden Open Variation Database (LOVD) platform using Apache, MySQL and PHP on Linux, Windows and Mac operating systems. This so-called LSDB-in-a-Box software requiring 15 min installation time is used primarily to store human variants, but can be used for other organisms. LOVD has been developed further within the Gen2Phen project to accommodate the collection of variants identified by next generation sequencing. 2) the Mutalyzer variation nomenclature checker based on a description of the standard human nomenclature syntax in EBNF format. Although developed as a generic stand-alone tool capable of handling reference sequence records from any organism, Mutalyzer’s web services are used as curational tools by LOVD to enhance the quality of variant descriptions. Drs. Taschner and den Dunnen are experts in the standard human variation nomenclature and actively involved in designing consistent variant nomenclature for the most complex genomic rearrangements. 3) the Shared Variant Database, a high performance database storing all variants identified by next generation sequencing and their frequency. Applying filters on the variant frequency allows submitters to separate rare variants with potential harmful effects from the probably less important ones which are found in many individuals.
- Experimental life science technologies, e.g. genomics (incl. DNA and RNA ‘next generation sequencing’ applications)
- Human inherited disease
- Mutation detection
- Next generation sequencing (NGS)
- Mouse models
- Gene expression analysis
Expertise and Track Record
The Shared Variant Database collecting whole genome and exome variant data identified by next generation sequencing projects in participating centers is unique in the Netherlands and an example for the world. Access to variant frequency data is only possible when submitting your own data. The LOVD platform is used by >95% of human gene variant databases world-wide, leading to database standardization and increased familiarity of the interface to the users. No other platform has an API supporting the same level of data interoperability. The registered LOVD databases contain >2.5 million variants in >170,000 individuals in 22,000 genes. The API supports access to the majority of them. Data from LOVDs can be shared automatically with genome browsers and dbSNP. Mutalyzer is unique in its capacity to check variants described in HGVS format from any organism or reference sequence (plants, fungi, etc). Its modular design supports software quick updates when guidelines or reference file formats change. Several commercial packages (e.g. Alamut, Gensearch) use Mutalyzer either as a reference to check the performance of their tools or to complement their functionalities.
1) The LOVD team led the Genetic Databases Work Package of the Gen2Phen project aiming for globally accessible variant data in a standardized, universal format to enhance database interoperability and standardization. One of our deliverables was the creation of > 1200 foundational gene variant databases for all genes involved in Mendelian disease (www.lovd.nl/Mendelian).
2) Many static gene variant database sites have been moved by us to the LOVD platform. This also involves reformatting data, selection of appropriate (genomic) reference sequences, nomenclature checks, data quality control, database customization and the setup of new gene variant databases or merger with existing ones. At the request of clinical genetics labs, we have imported around 2 million variants in 22002 genes from the Exome Variant Server after conversion into an LOVD database. We have assisted the creation of population- and ethnic-specific LOVD databases supporting the activities of the Human Variome Project Country Node initiative (See FinDis.org for example).
3) Summary data containing 28,000 variants (4200 unique) on chromosome X identified in a whole exome sequencing project involving 208 families with mental retardation (Tarpey et al, 2009 PMID: 19377476) have been imported after an HGVS variant description check into the Mental Retardation database (www.lovd.nl/MR).
Our LOVD development work was funded by the European Gen2Phen project. LOVD has received the Recommended System status from the Human Variome Project (www.humanvariome.org), which aims to convince all researchers and clinicians to share variant data openly. Team members participate in the NBIC Bioassist program. Issues regarding reference sequences and variant data sharing with Ensembl, dbSNP and Uniprot are solved in collaboration with EBI, NCBI and SIB. To enable broad support, the team interacts and collaborates with other participants in the Bioinformatics Research Group (See 5.2b).