[Updated: 9:20 pm Pacific, 3/10/10 with independent expert comment] Doctors and researchers in the future won’t just want to look at your genome to see how to treat or prevent illness. Instead, they will look at how all of an individual’s DNA compares with the closest members of their family, according to biotech pioneer Leroy Hood.
That’s one of the big conclusions from a study by Hood and his colleagues at the Institute for Systems Biology, the University of Washington, the University of Utah, and Mountain View, CA-based Complete Genomics. Researchers at those organizations studied the whole genomes of a family of four, searching for clues into why an otherwise healthy mother and father produced two children with a couple of rare genetic disorders. The findings are being published in this week’s issue of Science.
“The sequencing data from families is far more accurate than with random individuals,” Hood tells me. “This increased accuracy makes it easier to find the genes that caused these genetic diseases.”
Gene sequencing has been on a torrid pace of innovation over the past few years, as established toolmakers like Illumina, Life Technologies, and Roche have been racing to lower the cost of sequencing an entire human genome to as little as $10,000. Complete Genomics has raised the ante even more, saying its new business model allows it to sequence entire genomes for as little as $5,000. This better, faster, cheaper technology is making it possible for the first time to do studies like this in a family of four people, to find the underlying genetic abnormalities for a disease like Miller Syndrome, which causes facial malformations like cleft palate.
The new paper represents just the beginning of this kind of research, Hood says. The Institute for Systems Biology has also paired up with Complete Genomics to sequence 100 genomes of individuals, family members, and control subjects, to better understand the roots of Huntington’s disease.
“There’s going to be a flood,” of family genome studies, Hood says. “An enormous number of people with next-generation sequencing machines are going to do things like this.”
Gilbert Omenn, the director of the University of Michigan’s Center for Computational Medicine and Biology said, “this is a landmark paper.” Omenn wasn’t involved in the project.
By sequencing the genomes of a family, researchers were able to identify 70 percent of the sequencing errors along the 3-billion letter strings of DNA, because they were simply inconsistent with inheritance patterns first articulated by Gregor Mendel. When researchers sequence random individuals without having family members as a reference point, it’s much harder to find the errors, Hood says. It’s basically about finding a better way to separate the signal from the noise in a pool of data that’s as vast as the genome.
Sequencing the entire genomes of this family of four enabled researchers to narrow down their candidate list to just four genes thought to be primarily responsible for Miller Syndrome and primary ciliary dyskinesia, a progressive disease that affects the lungs and other organs that depend on cilia.
By sequencing an entire family, researchers were also able to see how much the genome changes from one human generation to the next. The researchers found that gene mutations from parent to child occurred at half the most widely expected rate.
“Our study illustrates the beginning of a new era in which the analysis of a family’s genome can aid in the diagnosis and treatment of individual family members,” said David Galas, one of Hood’s colleagues at the Institute for Systems Biology, in a statement. “We could soon find that our family’s genome sequence will become a normal part of our medical records.”
[Updated: 9:20 pm Pacific, 3/10/10, with comments from Gilbert Omenn of the University of Michigan]
Xconomy: What implications you think this publication might have for the growing field of complete genome sequencing, now that the price is coming down to a practical point to do studies like this?
Gilbert Omenn: This study will give complete genome sequencing a huge boost, especially as the cost declines and further applications are demonstrated. These kinds of studies will also put a premium on bioinformatics.
X: What do you think this approach is really useful for?
GO: This approach will have many uses, as already demonstrated in this initial paper. Basic questions about mutation rates in different sequence context or in varying regulatory context will warrant investigation. Searching in families for evidence of mutations of various types attributable to specific environmental exposures might advance our understanding of toxicology from chemicals and from radiation. Finding the molecular cause for rare disorders in analogous family quartets should be facilitated as costs drop. Recognizing sequence errors and removing them should become a routine part of sequencing studies.
X: What are its limitations?
GO: Repetitive DNA sequences, as in duplications and satellites, remain quite challenging.
As reported here, certain findings require extensive confirmation or validation with re-sequencing or orthogonal complementary methods, illustrated here with mass spectrometry. Data management and advances in bioinformatics tools will be at a premium.
The discovery of the molecular lesion for Miller syndrome still depended on study of two additional affected individuals, who could have had different alleles or even different gene loci involved.
