ASHG 2010 is now over, and I am back on Albion. Either me, or Daniel, or both (or, indeed, any of the other GNZers) will have personal genomics roundup over at Genomes Unzipped sometime this week.
For the last of these posts, however, I thought I would just report on the entirety of the Exome Sequencing session on the final day of the conference. I loved this session for the diversity, the number of different projects that are using exome sequencing to address old questions. It shows how much biology is tech-limited: the moment a powerful new technology becomes available at a low price it is used in every field by a flood of researchers who have been waiting for exactly this sort of data.
Other than that, there wasn’t an overall theme to the session (or to this blog post), other than Exome Sequencing Is Cool.
Doing it with exomes
Stacey Gabriel gave a lowdown of her sequencing center, which also acted as a pretty good guide to the logistics of exome sequencing. She estimated that exomes are now roughly 1/7th the price of whole genomes, and guessed that whole-genome sequencing would take over when it fell to less than 3-times the cost. Until then, she presented some exome vs genome sequence data off the HiSeq, and showed that 95% of variants were called by both methods, with the number of coding variants unique to each method being roughly equal (and enriched for false positives, I’d guess the number of true variants called by both methods is closer to 97%). She also made the interesting point that higher coverage of exome sequencing is particularly useful in cancers with high levels of aneuploidy, which can increase the depth required to detect heterozygous mutations.
Rick Lifton reported on a bunch of exome sequencing in Mendelian genetics, including some of medically interesting case-studies that lead to new diagnoses and possible treatments. My favourite example, published earlier this year in Science was the mapping of the gene for Ichthyosis with confetti, in which a patient has dry, flaky skin over most of his body, but with hundreds of obvious, small flecks of healthy skin. The disease is caused my a dominant gain-of-function mutation in Keratin 10, which is then secondary lost via a regional loss-of-heterozygosity due to mitotic recombination; over a period of years, the healthy spots get larger and more frequent as selection favours the healthy tissue. Rick Lifton cautiously suggested that a (potentially risky) treatment that increased the rate of mitotic recombination could speed up the rate at which the healthy tissue appears.
Jay Shendure took profound autism patients with no family history of autism, and sequenced the exomes of them and their parents to look for new, de novo mutations (a process he described as looking for needles in haystacks with less than 1 needle each). He found predicted protein-disrupting mutations in GRIN2B, FOXP1, SCN1A and LAMC2. Interestingly, de novo mutations in GRIN2B and FOXP1 have previously been implicated in candidate studies of intellectual disability: Hamden et al found loss-of-function mutations in FOXP1 in two out of 355 patients (and none in 570 controls), and Endele et al found GRIN2B mutations in 4 out of 468 patients (and again absent in a few hundred controls). The numbers in these studies were small, and not very significant given the rarity of the events, but Jay Shedure’s replication in his study makes them look far more plausible.
Gonçalo Abecasis gave a talk that did not actually involve exomes, but instead whole-genome low-coverage sequencing of 1000 Sardinians, with measurements of a number of metabolic traits involved in healthy aging. The variants discovered were then imputed into a further 5000 genotyped samples; this is really “squeeze every drop” genetics, trying to get genome-wide variation in 6000 individuals for less than the cost of 200 deep-coverage genomes (or about 1400 exomes). Most of the work he presented was on fine mapping; in particular, he noted a number of secondary signals at known loci, including a number that seem to be unique to Sardinia. To add to the Missing Heritability discussion: fine-mapping increased the amount of variance explained in LDL by known loci massively, from 3% to 7%.
Kun Zhang gave an original presentation on exome sequencing induced pluripotent stem (iPS) cells, and the fibroblast progenitors they were made from. He reported that each iPS cell line has around 6 new protein-coding mutations; worryingly, these variants were under strong selection, and were enriched for cancer-associated genes. This is suggesting some cancer-like behaviour in these cell lines (as you would expect, given unchecked selection for growth), and is possibly more worrying in light of a recent paper showed that iPS cells are more likely to develop into teratomas than embryonic stem cells. However, maybe these variants will also provide ways of lowering the risk of iPS-induced cancer either by screening lines, or by illuminated the basic biology.
