By comparing the genomes of vertebrates, a group at Washington University found that the typical human genome probably carries between 800 and 900 harmful mutations, many of which are rare in the population. The idea of this research was to develop a method that will look at your genome sequence and predict which of your many genetic variants are likely to be harmful. The prediction method is still fairly primitive at this point, but multiple lines of evidence indicate that we all probably carry 100's of rare, harmful mutations.
A group from Arizona State also has some results in the same journal issue. They also have a method for predicting harmful mutations, and this method also relies on sequence conservation - that is, the method is based on information we get by taking evolution into account. Evolution isn't just something to irritate Biblical fundamentalists - it's a powerful source of information that's essential to making sense of our genomes.
So how close are we to looking at your personal genome sequence and telling you which genetic variants are likely to be deleterious to your health? Not very - these methods aren't yet ready for commercial prime time, but impressive progress has been made.
What I find most interesting (if somewhat obvious with hindsight), is that we all carry around a significant batch of potentially harmful mutations. Most of these are recessive anyway, with small if any effects on your health. Which means that there is another hurdle to overcome after we get better at predicting which variants are potentially harmful: which ones are actually harmful, that is, which are recessive, which ones have negligible effects, and which have large effects.
These types of questions are tackled in another paper in this issue of Genome Research, written by a group at The Scripps Research Institute. Recently cancer genomes have come in for some intensive sequencing, and now the challenge is to identify which mutations in cancer cells actually contribute to the progression of cancer:
Analysis of the frequency of specific mutations across different tumors has been able to identify some, but not all of the mutated genes that contribute to tumor initiation and progression. One reason for this is that other functionally important genes are likely to be mutated more rarely and only in specific contexts. Thus, for example, mutation in one member of a collection of functionally related genes may result in the same net effect
The researchers used a network analysis method to suggest that some key cell signaling pathways (all known already to be involved in cancer, with exciting names like Wnt/TGF-beta cross-talk, Wnt/VEGF signaling, and MAPK/focal adhesion kinase pathways) are especially ripe for mutations that drive cancer progression. What's interesting is that the network model suggests that cancer mutations don't damage the core players in the pathways. Instead they:
contribute to a more refined shaping or “tuning” of the functioning of these pathways in such a way as to result in the inhibition of their tumor-suppressive signaling arms, and thereby conserve or enhance tumor-promoting processes.
This proposal is very testable (you know I hate untestable/untested network analysis). It also indicated that we need to understand how these key cell pathways are tuned, and that means building mathematical models. To understand cancer, we need to understand how cellular signal processing works as a system - a parts list isn't enough.
Read the feed: