Third day down, one to go. I am starting to suffer from conference fatigue somewhat. I’m not going to any other talks this evening, so I am going to try and get some relaxation time in from this point on. But first; the summary of Day 3.
Today I saw a lot of talks over three sessions, and many of them were very interesting. However, I won’t talk about everything, or even my favourite talks. I’ll go for the talks that seem to tie together into nice stories about a few directions genetics seems to be heading in.
Tracking Down Mendelian Mutations
A couple of people gave talks about tracking down Mendelian disorders. Matthew Bainbridge talked about how to track down dominant genetic mutations, given limited sequencing capacity; the answer is to carefully pick which individuals you sequence and validate, and in what order. Clesson Turner reported that about 2% of a group of heart disease patients had previously undiagnosed monogenetic disorders; they also predicted 25 relatives who were at risk, and 8 of them have been scanned, including one who had a serious monogenic heart disease without even knowing she had a heart problem.
This all ties in with a talk I saw last night by Evan Eichler, on tracking down the mutational causes of developmental retardation. Monogenic diseases may account for quite a lot of disease, when taken as a group, and finding next generation methods of quickly and cheaply tracking down the cause is valuable. Evan pointed out that, even if we can’t cure these diseases, just being able to bring families who suffer from the same disease together can allow you to create support networks, especially when, as in diseases like developmental retardation, it is easy for parents to feel alone.
More Inclusive Population Genetics
High-variant-density population genetics has long been the domain of the very white, the very East Asian and the very Yoruban. A couple of talks I saw today took high-density genotyping to other groups that have been understudied.
Andres Metspalu presented his collection of 4000 individuals from 21 European populations, most of which were Eastern European, massively improving the resolution in that region. Interesting lesson: despite having a very similar language to Finland, Estonians are far more closely related to their neighbouring Latvians than the Finns.
Li Hao talked about the Jewish HapMap Project, a large dataset of Jews from seven major Jewish populations. Each Jewish group was separately identifiable, though they did all show a distinct, Middle Eastern ancestry.
This was all bought together by Jinchuan Xing, who has collected genotypes from 860 individuals over 40 populations, with an attempt to spread these all over the world, including previously understudied areas like South Asia. One slightly heartwarming part of the talk was seeing the tightly racial clustering of the HapMap2 populations break down into a continuous, meandering trail of variation as more of the world was included. One interesting discovery was that Tamil and Andhra Brahmin cluster separately from the other castes, despite coming from very different parts of India.
There is something very cool about seeing genotyping data getting cheap and easy enough to collect that we can start seeing these diverse studies starting to come together; it is very easy to get lost in the big picture in genomics, and it is great to see individuals stories start to crystallize out of the masses of data.