Gene expression in the human brain

点击翻译按钮获取中文摘要

We have previously reported the generation of a large, integrated dataset whereby we, and other groups, can examine the relationships between genetic diversity and gene expression in the human brain. This has most immediate impact for understanding gene variants identified in genome-wide association studies where most nominated polymorphisms cannot immediately be assigned a function, as most do not change protein coding sequences. Rather, many are associated with differences in gene expression. We and others have used our data to derive such expression quantitative trait loci (eQTL) and have found them to be very helpful in understanding the genetic basis of a number of neurological and psychiatric conditions. However, our current dataset was generated using microarrays, which is a probe-based technique for estimating gene expression levels. Some of the known limitations of microarrays include that probes have only a single sequence whereas in the human genome, many genes are variable. Also, genes are alternatively spliced and edited which are poorly represented on most arrays. To overcome this, we are currently replacing our microarray based dataset with RNA-Seq, a newer technique that directly sequences expressed genes as well as providing measures of alternate exon using (ie splicing). We first applied this technique, and associated analytical approaches, to the mouse brain where we found substantial changes in gene expression, splicing and editing during development. Ongoing work in the laboratory includes applying the same approach to mouse models of disease and to a large series of human brains whose DNA has also been sequenced.