Tuesday, May 31, 2011

BGI Cognitive Genomics Unit

This is my first visit since the official creation of the Cognitive Genomics Unit (CGU). A few photos below.

Technicians work on Illumina HiSeq 2000 machines. These things go for $500k each.

Among the bioinformaticists. This building used to be a shoe factory :-)

Back at the hotel.

BGI: ... In May 2011, Nature Publishing Group published Nature Publishing Index (NPI) 2010 China, delivering “a dramatic rise in the quality of research being published by China”. BGI Shenzhen, the headquarters of BGI, is ranked 4th of Top 10 Institutions in the index.

Published as a supplement to Nature, the 2010 Index for China ranks research institutions and cities in mainland China. The ranking is based on outputs in Nature research journals in 2010 with comparative data for 2009. The supplement also presents data from other leading journals – Science, Cell, NEJM and The Lancet – showing similar rises in quality output from China.

( http://www.natureasia.com/en/publishing-index/china/2010/ )

NPI 2010 China describes BGI Shenzhen as "Taking the world by a storm". In 2010, BGI Shenzhen has contributed nine articles (CC 3.572) to Nature journals, with the majority of these in genetics and biotechnology. BGI Shenzhen is ranked second and third in the Asia-Pacific region on contributions to Nature Biotechnology and Nature Genetics. ...


David Coughlin said...

I had to look it up.  You have probably covered this and I forgot, but it makes me chuckle that the Beijing Genomics Institute is 2000 miles away from Beijing.

steve hsu said...

It's now just called BGI (formerly known as Beijing Genomics Institute). The Shenzhen govt lured them here with grant money, land, etc. Now other cities are trying to lure them away from Shenzhen.

MtMoru said...

Is there some theorem that the PC determined g should be more heritable than any other mapping of subtests to a single parameter? It sounds silly but ...

Why not merely seek that linear combination of subtests or perhaps nonlinear mapping of subtests to one parameter which maximizes heritability.

It would be no more arbitrary than using the dot product of an individual's subtests with the PC unit vector to assign g to indiviudals.

It is off topic, but has this occured to anyone?

Blog Archive