| |
|
Wasn't sure if this one deserved it's own thread, but I recently attended a very interesting talk on variation in copy number within the human genome. This was the first time I was exposed to something like this, and I'm having trouble finding a single link that explains it well, so bear with me.
As per GCSE biology, you have 2 copies of the vast majority of your genes, one from each parent, each on their own copy of a particular chromosome. But it seems there are hundreds, if not thousands of sites within the genome where this is not neccessarily true. Within these sites, you may get copies ranging from a single exon to multiple copies of the whole gene (which may differ allelically from each other). And of course there is no reason to expect the same copy variation to exist on both chromosomes. So for example you may have a gene with three copies of exon 5 on one chromosome, and a 'normal' counterpart, or eight copies of a gene on one chromosome (with different alleles), with no counterpart gene on the other. These variations are reasonably common, are thought to occur during meiosis (therefore hereditary) and may cover around 12% of the human genome (though they are present in animals, plants and fungi).
Obviously this has kicked off a lot of research into it's implications for disease. One of the early papers on copy number variation (Sebat et al., 2004) indicates that over 300 disease-related genes contain potential CNV sites, and a later report expanded that to nearly 1/6th of disease-related genes. Since these publications a number conflicting reports have started emerging on various conditions - e.g. whether variation within the gene CCL3L1 is thought to impact on HIV succeptibility or not (latest paper says not). Copy variation is also thought to impact on both the risk of cancer, and play a role in its progression (oncogene amplification is now thought to be extremely common in human tumours at least).
But what drives them? Obviously the majority of these variations can't have a huge impact on the risk of most diseases, or they would be selected against. But it is also possible that, as with e.g. sickle cell anaemia or CF, there may be conflicting selection pressures at work. Furthermore, many of these variations occur in introns and other non-genomic regions, the implications of which are anyone's guess. What is clear though is that this further undermines the Celera version of the human genome, increasing the potential variation between individuals massively, and showing that our genome is a lot more elastic than previously thought.
Thoughts? |
|
|
