The DECIPHER database - revolutionising clinical practice and genetic research

Apr 15 2009

Today - five years after the inception of the DECIPHER database - researchers have published a report that reveals the developing role of the database in revolutionising both clinical practice and genetic research.

The report explores the growing benefits of DECIPHER for researchers, clinicians and patients - highlighting how the data, provided by around 100 centres and shared openly worldwide, can benefit all three groups.

DECIPHER - the Database of Chromosomal Imbalance and Phenotype in Humans using Ensembl Resources - is hosted at the Wellcome Trust Sanger Institute. It was established in 2004 to catalogue submicroscopic structural duplications, deletions and rearrangements in the genome - called copy number variants (CNVs) - and to uncover their possible role in disease.

"The first comprehensive map of human copy number variation was produced just three years ago, changing our understanding of human genetics" explains Nigel Carter, a lead member of the DECIPHER team from the Wellcome Trust Sanger Institute. "Since then, over 10,000 CNVs have been found, covering about 5 per cent of the human genome. This rate of advance has been remarkable: using new technologies, we are able to uncover the smaller, elusive variants at a 50 fold-higher resolution. But the pivotal role that DECIPHER plays is in looking at how these variants affect human health."

The problem researchers face is that while many CNVs initially appear to have no visible effect on individual health, others appear to have minor effects, and some are harmful. What DECIPHER helps clinicians to do is to evaluate CNVs and determine whether or not they are linked to the patient's problems. In some cases, the findings are novel or have been observed only a handful of times before. With consent from the patient, data can be shared worldwide and clusters of people with overlapping genetic rearrangements can be identified.

By looking at genetic information first in an unbiased and less subjective manner, recurrent genetic changes can be found, researchers can then seek matching symptoms. This reverses the traditional practice of identification where researchers would move from individuals with shared symptoms back to a chromosomal cause and is particularly helpful for conditions such as learning disability and congenital disorders which have a large number of different genetic causes.

"We need new ways to uncover those rearrangements that cause human disease. But we must also be wary of dismissing CNVs if they appear to have no physical effect," says Charles Lee, an Associate Professor at Harvard Medical School and a Clinical Cytogeneticist at Brigham and Women's Hospital in Boston, USA. "For example, there may be variants that only affect people with a specific genetic makeup; or sometimes specific combinations of variants may result in pathology."

The report provides case studies in which DECIPHER played a pivotal role. In one example a four-year-old girl with symptoms of developmental delay and poor eye contact had a novel genetic finding and remained without a clear diagnosis. However, two new cases with similar structural variants were submitted to the database one year later, to provide the elusive diagnosis. The case studies exemplify increasing value of the database as clinicians add case information.

"DECIPHER is particularly useful when we look at patients with developmental delay, learning disability, dysmorphic features or congenital abnormalities, where, using genomic array technology, we can assign a diagnosis in 15 per cent of previously undiagnosed cases," explains Helen Firth, Consultant Clinical Geneticist at Addenbrookes Hospital and lead author on the study. "This improvement is dependent on a fantastic level of collaboration. More than 2000 patient cases have been contributed to the DECIPHER database since its inception: its diagnostic power strengthens as new cases are added"

DECIPHER is built upon the Ensembl genome browser. It is the only open-access, web-based interactive database of its type, although data from other databases are available. The report's authors suggest that while combination of all data in one resource would be ideal, providing access to the data in one genome browser is a realistic and practical method of harnessing the combined power of the datasets.

Sharing data between researchers is increasingly important. As the role of CNVs in human disease is better understood, so resources such as DECIPHER will gain momentum that will drive significant health benefits and improvements to genetic counselling.