Human genetic bi-allelic sequences (HGBASE), a database of intra-genic polymorphisms.

Abstract

The Human Genome Project is providing awealth of information about the human gene rep-ertoire, and promises to furnish a complete genomesequence (and thereby a complete gene catalog)by the year 2005. This enormous output of data isbeginning to be complemented by large scale stud-ies designed to uncover normally occurring varia-tions within human gene sequences. Much of thisvariability is very subtle, often comprises singlenucleotide polymorphisms (SNPs) which are ide-ally compatible with a number of large scale de-tection procedures. SNPs will be the basis of fu-ture highly dense polymorphic marker maps, andthose related to known genes can be exploited ingenetic association studies aimed at defining thegenetic basis of all manner of complex phenotypes,not least disorders such as mental illness, diabetes,cardiovascular disease and cancer. All indicationsare that 100,000-200,000 human genome SNPs willbe identified within the next two years.In light of the above developments, a databaseof gene based polymorphisms is obviously re-quired. To fulfill this need we have constructed andrecently released at http://hgbase.interactiva.de theHGBASE (human genic bi-allelic sequences) da-tabase of intra-genic sequence polymorphism.HGBASE is the result of a joint venture betweenUppsala University Medical Genetics Department,the Swedish Genome Research Centre, andInteractiva Biotechnologie GmbH. Its primary pur-pose is to facilitate genotype-phenotype associa-tion studies based upon the rapidly growing num-ber of known, gene related, single nucleotide poly-morphisms (SNPs) and other intra-genic sequencevariations. Furthermore, HGBASE will help to-wards the production of a dense SNP map of thehuman genome, which itself will be a valuable re-search tool.HGBASE is not designed to include gene ‘mu-tations’, but instead is a catalog of intra-genic (pro-moter to transcription end point) sequence vari-ants found in ‘normal’ individuals. Although thedistinction between ‘mutation’ and ‘variation’ canbe somewhat blurred, the general idea is that thecontent of HGBASE concerns frequently occur-ring ‘normal polymorphisms’, whether or not theyare suspected to increase the risk of developing aparticular phenotype. This is in contrast to ‘mu-tant sequences’ which are known to cause geneticdisease. Despite its name, HGBASE contains alltypes of intra-genic variation and is not limited tobi-allelic polymorphisms (though these do repre-sent most of the database content). Both functionalpolymorphisms (e.g. promoter and non-silentcodon changes) and non-functional polymorphisms(e.g. intron sequence differences) are included. Thisis for two reasons. Firstly, it is often difficult to becertain about the functional consequence of a varia-tion. Secondly, regardless of functional relevance,any intra-genic polymorphism can usually be em-ployed as an effective surrogate marker for an un-known functional variant in an association study,due to close proximity and linkage disequilibrium.Gene polymorphisms may be retrieved fromHGBASE by using the database search facilitiesto query either by a text string or by a DNA se-quence. Data submission to HGBASE is madesimple by provision of a series of Web page datasubmission forms. All submitted data is made avail-able to any other public database that wishes todownload it, and continual efforts are made to ac-cess new relevant data from other databases andliterature publications. The exponential growth inpolymorphism discovery requires that scientistsmake every effort to submit their data to theHGBASE database to ensure it remains up to date.HGBASE does not claim any rights to publiclyavailable or submitted data, instead this remainsthe property of the original submitter. Depositionof data into HGBASE requires only the allelic DNAsequences, the allele frequencies, the host genename, and the intra-genic domain. Additional com-