Friday Links

We’ve been pretty quiet here at Genomes Unzipped for the last couple of weeks, while many of us prepared for and attended the American Society of Human Genetics meeting in Washington DC last week. The meeting was (as usual) an exhausting barrage of presentations on human genetics; you can get a flavour of the new results presented either by scrolling back through the Twitter coverage, or reading the summary posts by Luke Jostins (who found himself explaining why Eric Lander was wrong about epistasis and missing heritability) and Larry Parnell. Shirley Wu also has an excellent recap of the meeting, including an analysis of the Twitter coverage.

My overall impression of the meeting – and this was a view shared by many other attendees I’ve spoken to – was  that 2010 was a year of transition. The numbers were impressive: Stacy Gabriel at the Broad Institute blithely reported that their pipeline is now theoretically capable of sequencing up to 800 exomes (complete sequence of the protein-coding regions of the genome) per week, and the 1000 Genomes Project has now completed sequencing of over 1,000 whole genomes at low coverage. Projects that would have seemed mind-blowing even a year ago seem almost mundane now, an indication of just how fast sequencing technology has moved in the last twelve months.

However, there was also a sense of frustrating incompleteness to many of the presentations: sure, we’re sequencing thousands of human exomes and hundreds of genomes, but we’re still not entirely sure exactly what we need to do to extract the most useful information from all that sequence data. Exome sequencing has so far resulted in the unravelling of at least 10 Mendelian diseases, reported Jay Shendure – but talking to people at the centres doing the most exome sequencing, I heard of many cases of apparently simple recessive diseases where a clean answer remains elusive even with near-complete protein-coding sequences of multiple family members in hand. The picture is even more complex for more common, multifactorial diseases.

In some cases the problem will simply be analytical, and I’m optimistic that by ASHG 2011 we will have a far better sense of how to handle these data (I hope to have a post up soon laying out some of the key challenges ahead). Still, the overall attitude at the meeting this year seemed rather cautious and thoughtful – it’s now clear that simply blasting biological problems with ever-increasing sequencing capacity won’t (in most cases) provide simple answers, and there will be plenty of work for clever analysts over the next twelve months.

There was also plenty of material at the meeting pertinent to personal genomics – I’ll have a post up next week summarising this. [DGM]

One of the plenary talks at ASHG was given by Jo Knight from King’s College London who was discussing a genome-wide association study of psoriasis that is reported in this month’s edition of Nature Genetics. As well as identifying eight new psoriasis risk loci they also discovered something that has been somewhat elusive to complex disease geneticists – a gene-gene interaction. Not surprisingly, given that psoriasis is an autoimmune disease, the interaction involves the HLA – an extended locus on chromosome 6 containing many genes involved in the immune system. The other gene involved is ERAP1 on chromosome 5, which is itself associated to psoriasis risk. The authors found that the ERAP1 locus is only associated with disease in individuals carrying the HLA-C risk variant. This finding is also interesting from a biological standpoint because ERAP1 regulates and fragments small peptides prior to their recognition by HLA class I molecules such as HLA-C. [CAA]