Blog @ Illumina
Real scientists. Real commentary.

Illumina Discovery Symposium at ASHG, Part Two

by
Abizar Lakdawalla
| Nov 14, 2012

photo courtesy of Otzi's facebook page: https://www.facebook.com/OetziTheIcemanAfter a great lunch, Carlos Bustamante from Stanford gave a phenomenal presentation (as usual) though his typically loud presentation voice seemed a bit moderated. Maybe it is the new beard he was sporting, which may help absorb some of the sound waves.

Using migration maps from 150k to 40k years ago, Carlos talked about human origins and the sharing of common and rare alleles in different human populations and discussed the value of genotyping-by-sequencing, akin to reduced-representation sequencing. From the Thousand Genome Project (TGP), most rare variants remain “private” to populations, with not many shared alleles. He showed a table with the cost effectiveness of different sequencing approaches including exome sequencing, genotyping-by-sequencing, and whole-genome sequencing. Genotyping-by-sequencing offered a good balance of low cost, genome-wide data coverage, and 99.9% concordance with Complete Genomics whole-human data. In an Argentinian population, sequencing-by-genotyping of SNVs found showed a 99% concordance. Carlos then talked about the sequencing of the Ötzi iceman. Data obtained from the SOLiD platform showed that Ötzi might have been a Sardinian, possibly on an extended vacation into the Alps. Comparing Mr. Ötzi with population data from TGP and from Complete Genomics showed equidistance from modern populations, including Tuscans.

So, did initial sequencing evidence confirm that Ötzi was a Mediterranean tourist? Carlos left us with this cliffhanger (no pun intended!), which was answered later at ASHG through the work of Martin Sikora of Stanford. Sikora sequenced Ötzi’s genome and compared it to several other pre-historic genomes, confirming the link to present-day Sardinians, and also found he was more similar to pre-historic farmers in Bulgaria and Sweden than to other hunter-gatherer groups for which sequenceable material exists. These findings support the hypothesis that human migration patterns from the Middle East to Northern Europe were impacted by farmers. The relative isolation of populations such as present-day Sardinians allows those genomes to retain greater percentages of Neolithic genes. Fascinating stuff, as were many other population genetics talks during ASHG.

David Goldstein from Duke University spoke of the challenges behind the sobering fact that up to 50,000 births per year in the United States are those with a genetic disease cannot be resolved by traditional methods. David is using sequencing to understand the de novo mutations behind novel presentations of genetic conditions. He discussed exome sequencing for childhood alternating hemiplegia (AHC) in seven trios, identifying common mutations in the sodium/potassium ATPase ATP1a3, confirmed by the sequencing of this gene in 91 additional patients. Mutations in specific regions had very different phenotypic effects. In AHC, mRNA and protein levels were not changed, but transporter function was affected. ATPase activity was decreased in a different condition known as abrupt-onset dystonia, in which multiple de novo recurrent mutations were found in the transmembrane region at the business end of ion transport. In a second example, three siblings with severe epileptic encephalopathy showed consistent mutations in the tyrosine kinase TNK2, and  functional assays showed interaction with EGFR.  In a third example, trios with an unreported phenotype had mutations in TCF4 that destroyed protein expression. Similarly, mutations in EFTUD2 in a case of microcephaly and in NGLY1 in a glycosylation disorder had no protein expression. An interesting cautionary note was that many of the sequenced controls also had what were called “damaging” mutations by at least one type of predictive algorithm.

Although I didn’t get a chance to write them up here, great talks were also given at the Illumina Discovery Symposium by Tom Royce, an Illumina bioinformatics scientist, Cisca Wijmenga from University Medical Center of Groningen , Erwin Bottinger from Mount Sinai, Michael Snyder of Stanford, and Gonçalco Abecasis from the University of Michigan School of Public Health.

Comment

  1.