Something interesting in my mail today:

A British bioinformatician wants to open discussion about the greatest biological discovery made possible by bioinformatics. He has just launched a "brainstorming competition" with a small cash prize to stoke discussion. Kaggle is hosting the "brainstorming competition" at:
http://kaggle.com/blog/2010/06/17/%EF%BB%BFwhat-has-bioinformatics-ever-done-for-us


For me, bioinformatics hasnt made any *single* great discovery, but has actually enabled countless discoveries through advances in sequence comparison (blast of course), network analysis, microarrays, proteomics. All of these fields have been critical components of biological discoveries in the last 10-20 years which have built on these technologies.

Recently the Boone group published a short report concerning the generation of systematic gene deletion for another strain of yeast: Genotype to phenotype: a complex problem.


Its amazing how much work went into this short reports, but this is the first study to show that genes required for growth can be significantly different between closely related strains of the same species. The article from the Boone lab stems from the observation that the genes required for growth in one strain of yeast can be dramatically different from another strain. This strain-to-strain difference is surprising and suggests that separate pathways are required for growth among very similar strains through the evolution of mutations which can buffer these essential pathways. The lessons learned here will likely be reflected in humans where individual variation can contribute to dependence on different sets of pathways and genes. An observation which has relevance to how people think about drug discovery and personalized medicine.

Were at ICSB this year and heard about a very interesting paper from Tobias Meyer. Using an intricate assay for wound healing in 96 well plates and automated microscopy, Tobias Meyer's lab looked for signaling components which could effect would healing. In this very clever approach they were able to track single cells as they migrated in response to a wound on a glass plate, in response to various stimuli and under different siRNA knockdown targets. By tracking each cell and measuring intricate cellular parameters such as cell motility, directed migration, coordination, attraction, repulsion and pioneer cells (those which move without the neighbors moving) they were able to assign genes to several functional modules in terms of a specific cellular movement parameter. Using all this data they were able to uncover parameters to an equation which closely follows actual cell migration, and genes which can specifically effect certain of these parameters. Its a very interesting approach and certainly a huge step in moving from siRNA screening to building predictive models of cell function directly form this data.

This paper in Nature from Ian Taylor and Jeff Wrana from U. Toronto present a compelling method for the discrimination and diagnosis of disease states based on the disruption of co-expression patterns with neighbors in a protein interaction network. Typically, the search for cancer biomarkers involves searching for differentially expressed genes which alone can discriminate between different cancer diagnoses. The authors present a method which integrates co-expression and public protein interaction information to show that genes whose protein network neighbors become dysregulated may be used to predict breast cancer outcomes. Their findings support the emerging notion that biomarkers might be improved by focusing on pathways rather than single genes alone. A couple of recent papers from our lab indicate show early signs of promise in this regard. Their results have implications for the systematic discovery of markers in a variety of diseases.

Genetics meets social networks! In a paper today in PNAS Fowler et al compare the topological characteristics of the social networks of MZ and DZ twins and find that some network properties are heritable. This is expanding on some previous work that showed that the −G1438A polymorphism within the promoter region of the 5-HT2A serotonin receptor gene is associated with variation in popularity. The results are interesting, that among teenagers in a various scools, that the number of times a person is named as a friend is heritable (more similar for MZ twins than DZ twins). Adding to this there were another handful of features that were heritable. Overall, this topic is pretty thought provoking, i wonder how many parents would line up for a drug to increase popularity!

I just ran across an interesting paper in Nature on the genetic basis of tumor susceptibility and skin inflammation. The authors use backcrossed mice to identify eQTLs and mapped these onto a network for co-expression based on various skin expression profiles. The expression profiles highlighted skin specific networks such as keratins. Their analysis suggests to me that using a generic co-expression network would not have given them the results that a "local" co-expression network which is based on skin expression which is the system they are interested in. Overall the paper presents an interesting approach for applying networks to eQTL analysis

The-scientist.com released its 2008 salary survey today. There are some interesting tools in there to find out the average pay for professors of different rank in different states (Avg pay for tenured faculty in California is 150K). Somewhat disturbing is the observation that women at the same rank make significantly less than men (23% less pay for women full professors vs men).

Check it out: http://www.the-scientist.com/salarysurvey/

As most reports of this type tend to show that salaries are rising over time in the life sciences fields, how does this jive with reports that there are a glut of postdocs with fewer than 20% finding faculty positions? Isn't this breaking some supply-demand curves? Or does this just mean that theres a glut of inadequately trained PhDs? Maybe this is a job for freakonomics!