Rare disease registries and mutation/variation databases

Abstract

Data regarding the effects of gene mutations on patients and proteins have been kept for research purposes since the 1950s with the globin protein. Increasingly, many researchers have collected data within gene locus-specific databases (LSDBs) for research and clinical use [e.g., Claustres et al., 2002]. The Human Gene Mutation Database (HGMD; www.hgmd.org) has been collecting virtually all instances of each specific disease-causing gene mutation from the published literature, but not all reports of each mutation. Online Mendelian Inheritance in Man (OMIM; www.omim.org) only collects the first few described and the most interesting examples of disease-causing mutations. This activity is becoming more critical as recently indicated in Nature Reviews Genetics [Samuels and Rouleau, 2011]. The Human Variome Project (HVP; www.humanvariomeproject.org) was initiated in 2006 by representatives of the global community to “collect all mutation data from all genes affecting human disease from all countries” so as to ensure the delivery of optimum genetic healthcare. The importance of this collection of all instances has been underlined [Cotton et al., 2009]. Recently and in parallel, the field of rare diseases has been promoted [Watson et al., 2008] and as part of this, the importance of rare disease registries has been underlined [Groft, 2011; Forrest et al., 2011]. All this has culminated in the International Rare Disease Research Consortium (IRDiRC), which is a major step forward [Abbott, 2011]. These two organizations come from different directions, the HVP from the genetic labs and clinics and the IRDiRC from the rare disease community. Thus, there is now a critical need for collecting all instances of mutations worldwide (HVP) and all patients with rare diseases (80% of which are inherited; IRDiRC). Both these groups are committed to building their individual registries. It would seem wasteful if both were built in parallel considering parallel function, pathways and networks worldwide might be developed. The article in this issue by Van Den Akker et al. [2011] describes an online resource, which is not only critical for bridging these databasing strategies and needs in this area, but also provides an overall model. The resource, called The International Dystrophic EB Patient Registry (www.deb-central.org), is an online database of dystrophic epidermolysis bullosa (DEB) patients and their COL7A1 mutations. The most critical point is that this registry not only collects records of all patients worldwide, but also defines their mutations. It is clearly not complete yet, as it only has 70% of the mutations currently found in HGMD. Also, it is not stated whether or not all relevant labs/clinics worldwide feed data into the database, which would be critical for such a database/registry to be comprehensive. There is however a call for further collaboration, which is also on the website of Debra international (http://www.debra-international.org/epidermolysis-bullosa/eb-classification/), an umbrella patient advocacy group whose members are the national associations representing people with EB worldwide. Other critical points about this database/registry are that unpublished data are included; the software is open source; the clinical phenotypes have been standardized by consensus; referee sequences and mutation nomenclature have been adjusted to recommended standards; a curator enters published data; there is a news page; linkage of articles is encouraged; visualization is provided; provision is made for nonconsented entries; and the software has been converted to LOVD (www.lovd.nl) format, which is widely used (over 1,000 downloads; J.T. den Dunnen, personal communication, 2010). Having listed all these points, it seems hard to find an omission in what might be taken as an ideal model for the future in the HVP/Rare Diseases Initiative. The International Dystrophic EB Patient Registry seems very complete although, as unfortunately happens more often, information about the RNA molecule as the intermediate between DNA and protein seems not to exist. Users therefore have no clue whether RNA has been analyzed or not with these patients and whether there is any experimental evidence for the change reported at protein level. This is an omission that was missed by the reviewers as well. Naturally there may be other databases, such as TREAT-NMD database (www.treat-nmd.eu), which have many of the above features also, and the LOVD LSDBs also have the capacity to act as a registry. This solution is fine for diseases with frequency of around 1,000 worldwide, as in DEB, but for a disease like colon cancer (even for the inherited data [www.insight-group.org] let alone all cases), collection may need to be on a country-by-country basis with links or downloads to a central system. Thus, it is likely that this article by Van Den Akker et al. [2011] might be a watershed, at least in the inherited disease arena, and stimulate registry creation. Correspondence is welcomed on this topic.