Bioinformaticians love contests. There are competitions for how best to pick gene expression that signals disease, foretell how a protein will fold, assemble genomes and predict how mutations affect protein function. By comparing different approaches to solving the same problem, researchers can develop better solutions.
But the latest contest announced at the meeting for the American Society for Human Genetics (ASHG) is more serious than those, because the true prize is helping ill children and their families. Accordingly, contestants were judged not just on the accuracy of their answers, but also for their clarity and precision in explaining them.
This contest, dubbed CLARITY (‘children’s leadership award for the reliable interpretation and appropriate transmission of your genomic information’), challenged teams to pin down genetic causes for rare disorders found in patients cared for at Boston Children’s Hospital in Massachusetts.
Each CLARITY team received sequencing, inheritance and clinical data from three families in which children had mysterious disorders that had thwarted all previous attempts at diagnosis. Teams used informatics tools to find a gene or genes that might be responsible and produced a prototype report to show how results would be presented to patients’ physicians or families.
The contest follows clear trends in genetics. This year’s ASHG meeting is awash in studies using exome or whole-genome sequencing on human patients to find genes potentially responsible for devastating, hard-to-explain diseases. Strong genetic contributors for common diseases are still elusive.
Participants in the CLARITY contest found mutations in an 11-year-old boy that explained his muscle weakness and hearing loss. Eight teams found the gene responsible for the former and six found the one for the latter; three teams found both genes. One of the other cases was a baby that had died when it was two weeks old because of heart irregulaties: seven teams identified a mutation in a cation channel gene as potentially causative. The third patient has muscle weakness. In this case, teams did not home in on a particular variant but did identify several for follow-up studies, including genes not previously indicated in the disease.
Forty teams submitted credentials to CLARITY sponsors at Boston Children’s Hospital, who determined that 32 had sufficient expertise to participate. Ultimately, 23 submitted their answers and eight teams were selected for commendations, with top honours going to a team of computational biologists and clinicians at Brigham and Women’s Hospital in Boston.
Alan Beggs, director of the Manton Center for Orphan Disease Research at Boston Children’s Hospital, conceived and organized the contest along with Isaac Kohane and David Margulies. Beggs noted two aspects of the best-performing teams. They were more likely to be multidisciplinary than to consist primarily of bioinformaticians or clinicians. Also, they considered how the quality of sequencing data varied across the genome. The vagaries of the technology mean that the data accrued for some parts of the genome are better than for others. The winning teams cross-referenced data from different sequencing technologies and flagged certain regions for repeat tests.
Sarah Foye, the mother of the boy with muscle and hearing problems, showed reporters at the press conference a picture of her son. The contest has not revealed new treatment options, but after ten years of inconclusive tests, she said, knowing the cause of disease means much. “There isn’t necessarily a treatment or cure, but it is a new stepping point.”