By Kevin Davies
Oct. 8, 2008 | There was much activity at the Broad Institute last month. First there was a ceremony to mark the historic pledge of $400 million from billionaire philanthropists Eli and Edythe Broad to provide an endowment for the institute that guarantees its future (not that there was really any doubt—see page 8).
Two days later, 300 scientists gathered for the second Wellcome Trust/Nature Genetics symposium on the Genomics of Common Diseases. I’m not going to detail any of the highlights here—the meeting followed Gordon Conference rules designed to encourage open presentation and discussion of unpublished results. Suffice it to say that the rapid progress in defining the genes and copy number variants (CNVs) responsible for common diseases such as cancer, heart disease, diabetes, and mental illness continues apace.
The biggest splash in genome-wide association studies (GWAS) in 2007 was made by the Wellcome Trust Case Control Consortium (WTCCC), which identified dozens of gene variants for seven common diseases. In some cases, such as Crohn’s disease, more than 30 confirmed gene associations have now been reported. This summer, a WTCCC sequel was announced, in which consortia will use new technology from Illumina, Affymetrix, and Agilent to analyze 120,000 DNA samples. This project, considered the world’s largest genetic research study in history, will include both SNPs and CNVs in 27 diseases such as multiple sclerosis, asthma, and schizophrenia.
But simply compiling lengthy lists of genes using GWAS and resequencing targeted regions (see page 13) won’t be enough. For a start, GWAS only provides the power to detect relatively common gene variants. A growing consensus is that rare variants may cumulatively explain much of the unexplained genetic variance in common disease susceptibility. Indeed, several speakers at the Broad meeting presented examples of sequencing projects in which mutations of rare or intermediate frequency have been uncovered. The deCODE group in Iceland recently published a method that it believes will help discover rare disease-causing variants, at least until we get closer to the $1000 genome (Kong, A. et al. Nature Gen. Sept 2008).
The other challenge is how to translate lists of seemingly unrelated gene-disease associations into useful diagnostics or therapeutics. Few groups have made as much of an impact in this arena lately as Eric Schadt’s team at Merck (he’s part of the Rosetta Inpharmatics subsidiary in Seattle)—the subject of this month’s cover story (see page 20). Schadt’s team is leading a multidimensional approach to tackling common diseases.
Rather than just look at genetics or gene expression, Schadt integrates these data streams and others. As detailed in our story, in at least one published case, his team showed that one of the WTCCC disease-associated genes was likely to be incorrect. More importantly, Schadt considers the assessment and modulation of specific gene circuits rather than individual genes. (You can read John Russell’s entire interview with Schadt in the inaugural issue of Predictive Biomedicine, the latest Bio-IT World eNewsletter)
Several complementary approaches are also being tried to make sense of the GWAS bounty, including functional genomics methods that look at protein-protein interactions, as well as text mining of the scientific literature. Various methods of capturing coding genes and resequencing also offers exciting opportunities, as does improved understanding of the extent of CNVs.
Such methods will prove invaluable in unraveling the genetics of cancer. The latest tour de force analysis from Johns Hopkins’ Bert Vogelstein, Victor Velculescu, and Ken Kinzler (see page 34) have documented hundreds of mutated genes in dozens of pancreatic and brain cancer samples (echoing findings in breast and colon cancer reported in 2007). The average tumor cell carries more than 60 genetic alterations, with one gene mutation seemingly distinguishing different severities of brain cancer. Nevertheless, the authors conclude that the best route to new therapies probably lies in targeting “the physiologic effects of the altered pathways and processes, rather than their individual genetic components.”
BEST PRACTICES ’09: CALL FOR ENTRIES
Following the gratifying success of this year’s Best Practices competition (see August issue for profiles of the winners and summaries of all 56 entries), we are pleased to announce the kick-off of the 2009 contest. Full details, guidelines, and categories are posted online (see www.bio-itworld.com/bestpractices). We are seeking to identify and showcase outstanding examples of innovative partnerships, technologies, and strategies impacting research and drug development. The winners, as judged by a formidable expert panel, will be feted at the 2009 Bio-IT World Conference & Expo (April 28, 2009). What are you waiting for?!
This article appeared in Bio-IT World Magazine.
Subscriptions are free for qualifying individuals. Apply Today.