VANCOUVER, BRITISH COLUMBIA (GenomeWeb News) – At the American College of Medical Genetics annual meeting here this week, Illumina UK researcher Mark Ross discussed strategies that Illumina researchers are using to interpret information in genome sequences from healthy individuals generated through the company's clinical whole-genome sequencing service.
To begin annotating individual genomes with information of potential clinical significance, Ross explained, the team first hammered out their basic strategy using genome data for an anonymous Yoruban man who was sampled through the HapMap study and whose genome was previously sequenced at Illumina.
In general, Ross and his co-workers came up with five tiers for classifying genetic variants: those found in a single gene that apparently cause disease; pharmacogenomically informative markers; variants associated with a common, complex disease; previously undocumented polymorphisms in known risk genes; and tissue-type variants.
For their initial interpretation efforts, the researchers focused on tier 1 variants found in the Human Gene Mutation Database, which they further classified as "disease-causing," "probable" disease-causing, "possible" disease-causing, or "likely benign."
The HGMD resource contains some 108,000 entries on mutations affecting more than 3,000 genes, Ross noted, and roughly two-thirds of these represent single base changes.
When they classified single base substitutions in the Yoruban HapMap genome, the team narrowed in on a single homozygous, Wilson disease-related change within the ATP7B gene. Because phenotype information and health records were not available for the individual, who was a father in one of the Yoruban HapMap trios, Ross explained, they were unable to take this analysis further.
Next, they applied to same strategy to interpret genome sequence data for the first individual sequenced through Illumina's CLIA-certified Clinical Services Laboratory.
The individual's physician ordered this individual genome sequencing, Ross said. Following genome sequencing, annotation, and interpretation, findings from the analyses were given to the physician, who subsequently delivered the information to the individual in the presence of a genetic counselor and a scientific liaison from the sequencing lab.
Initial analyses of the genome sequence turned up 16 homozygous and 48 heterozygous alleles that initially appeared to be disease causing. Nearly all of the changes were ruled out as disease causing following further scrutiny, though, Ross explained, noting that a dozen of the homozygous variants and several of the heterozygous variants were also present in the Yoruban individual.
Along with their additional analyses, the findings point to low levels of "contamination" of HGMD mutations with common polymorphisms, Ross said, though the team did track down at least one autosomal dominant mutation tied to a familial cold autoinflammatory syndrome for which the sequenced individual reported having some symptoms.
Though there are still challenges to be overcome in individual genome interpretation to ensure clinical utility and validity, Ross expressed optimism about the prospect of developing standardized genome assessment and interpretation methods with guidance and consultation from others.
"Whole-genome sequencing holds tremendous potential for clinically important information throughout the lifespan," he and his co-authors wrote in the abstract for the presentation. "The development of interpretation tools is essential to assist the physician in understanding and management of the implications of genome information for the patient."