Sanger Institute - The nonsense in our genes

Uploaded by sangerinstitute on 17.11.2009

A SNP is a single letter change in our genetic code. So, our genome is made up of four letters, which are called A, T, C and G
And when you have a SNP there's a change in one letter to another
So, for example, you could have a mutation where an A would turn into a G
So that some individuals would have a G and others would have the A. So, you get variation at that site in the genome
So, a nonsense-SNP is when the mutation causes a disruption in the protein
So it causes a premature stop to the reading of the protein
So that some individuals will have the normal, active version of the gene
While others might have the inactive, truncated, or even lost version of the gene
We tend to think that adaptation will be acheived by genes that are somehow better
So that genes that have been disrupted of even lost - we tend to think that must be a bad and harmful thing for you
However, there are known examples from the literature that we've seen
Where losing a gene has actually been beneficial for the individual carrying it
So we decided to do an overall scan of these kind of things and we looked at nonsense-SNPs as a type of gene loss
And so we scanned all the available nonsense-SNPs in the genome in a set of 12,000 individuals
From populations around the world
For 99 of these genes, both copies could be lost in these individuals
That actually seem to be living a normal healthy existence
So when you think about the fact that the genome contains roughly 20,000 genes
This means that roughly one in 200 is possibly just dispensible from the genome
It doesn't really matter for us to lose those genes
We actually saw that Craig Venter has three times more inactivating variants than the average human in our samples
And you might take this to mean that there's a whole lot of nonsense going on in Craig, but that's not necessarily the case
But the fact of the matter is that his genome, in his case his whole genome was sequenced
Which would give you a much more in depth view of the variants in his genome
Whereas ours are only based from the databases, which are kind of sketchy and distributed across the genome
So, actually we expect that, for our nonsense-SNPs, all of the estimates that we do are lower bound estimates
So, no it's not surprising that he had a higher number
So actually, in general, the nonsense-SNPs that are known in the databases
And the ones we use in the study are only just the tip of the iceberg
And this will be changed a lot when new sequencing studies become available
When they're getting better and more importantly maybe cheaper
This will lead us to discover much more variants and especially rare variants
So we'll definitely find a whole lot more, a higher number of nonsense-SNPs in the human genome, actually
When we get these sequencing studies through
So in medical genetics, people are looking for disease causing mutations
And when they come across a mutation like these nonsense-SNPs, which actually disrupt
Or even - you know - completely inactivate the gene
They might really tend to hang on to the idea that this must be a harmful disease mutation
However, in this genome-wide survey of ours, we've shown that actually some of them can be lost in healthy individuals
Without any obvious consequences
So in those type of studies - in these kind of new personalised medicine and medical sequencing
It might be worth at least excluding these ones that don't seem to be having any effect on the individual's health
We find that there are certain types of genes that seem to be lost from our genome
And in fact we found quite a high number of ones that are involved in our sense of smell
So it's possible to, kind of, consider the fact that in early times humans didn't like smelly partners
So that, when humans started living in bigger groups, it was advantageous and perhaps the only way to find true love
That you couldn't really smell - you know - your partner. But that's of course a very wild speculation