Today’s tip is on a new database based on data from a single interesting paper, SNPxGE2. With a large scale association study from HapMap data (269 individuals, 4 populations, over 500k SNPs and 15k expression profiles), the research reported:
the computationally predicted human SNP-coexpression associations, that is, the differential co-expression between 2 genes is associated with the genotype of an SNP.
This data is organized in an easily searchable database called SNPxGE2. As the paper only came out 2 months ago, it’s a promising database. It’s interesting and helpful as is, but I can see more data being added over time.
Welcome to our Friday feature link collection: SNPpets. During the week we come across a lot of links and reads that we think are interesting, but don’t make it to a blog post. Here they are for your enjoyment…
Today’s tip is on a resource named SPELL, which stands for ‘Serial Pattern of Expression Levels Locator’. You enter a small set of gene names. The SPELL search engine will analyze expression datasets collected from GEO, ArrayExpress, SMD, etc. and tell you which datasets are most informative for your genes. It will also return a list of other genes with similar expression profiles. The results link to the original publications and to gene summaries at SGD, and provides you with a list of over-represented GO terms. I found SPELL through SGD’s Expression Connection (a wonderful resource in and of itself), and liked what I saw. Plus the name ‘SPELL’ is sort of appropriate for Halloween, which is fast approaching. Read more about the resource here, which is also referenced below.
Reference: M. A. Hibbs, D. C. Hess, C. L. Myers, C. Huttenhower, K. Li, O. G. Troyanskaya (2007). Exploring the functional landscape of gene expression: directed search of large microarray compendia Bioinformatics, 23 (20), 2692-2699 DOI: 10.1093/bioinformatics/btm403
So, let’s say you need to find genes that are not only highly expressed in a tissue of a species of interest, but predominately expressed in that tissue (not highly expressed in other tissues). There are, like with any question, several ways to go about it, but in today’s tip of the week, I’m going to show you how to do it using the Gene Sorter. This is a tool brought to you by the same people who do the UCSC Genome Browser (Golden Path). The basic purpose of the tool is to take a gene of interest and sort other genes (which ones to sort can be filtered) in the genome based on some type of similarity (name, GO terms, expression, protein similarity, etc). What we are going to use it for in this tip is a little different since we don’t have a gene of interest, we are looking for interesting genes. There is a free tutorial and training on the UCSC tools including the Gene Sorter (UCSC Advanced Topics, download the slides and view/read the “Gene Sorter” section and do the exercises for the advanced topics). Know of other ways to do this? Suggest them in the comments!
The Allen Institute for Brain Sceince is a great institution that was founded just under 5 years ago with a 100 million seed money from billionaire Paul Allen (of Microsoft fame). The purpose is,
… dedicated to performing innovative basic research on the brain and distributing its discoveries to researchers around the world. Through its efforts, the Institute aims to advance a new understanding of brain diseases and disorders.
The result of this research is disseminated through some excellent tools at the Allen Brain Atlas. This research and tool focuses on the mouse brain and determining which genes are expressed in different parts of the brain.
Well, it was recently announced that not only are they planning to extend this map to the mouse spinal chord and another atlas of brain development from fetus to adult mouse, they have launched a project to do a similar atlas of the human brain. This project is expected to take four years.
btw, the “brain explorer” tool is just cool. My expertise isn’t mouse or brain science, but I like roaming around the brain as much as the next guy :).