Are mutations truly random?
Do genetic mutations really occur at random spots along the genome, as researchers have long supposed? Maybe not, according to a study published online today (January 13) in Proceedings of the Royal Society B, which proposes a mechanism for how new mutations might preferentially form around existing ones.
Image: Wikimedia commons, Jerome Walker, Dennis Myts"The idea is quite interesting," said evolutionary geneticist linkurl:Maud Tenaillon;http://moulon.inra.fr/pages_pers/tenaillon/ of the Un
Do genetic mutations really occur at random spots along the genome, as researchers have long supposed? Maybe not, according to a study published online today (January 13) in Proceedings of the Royal Society B
, which proposes a mechanism for how new mutations might preferentially form around existing ones.
|Image: Wikimedia commons, Jerome Walker, Dennis Myts|
"The idea is quite interesting," said evolutionary geneticist linkurl:Maud Tenaillon;http://moulon.inra.fr/pages_pers/tenaillon/ of the University of California, Irvine, who was not involved in the research. "I think it could be a good explanation for [mutational] hotspots." But, she cautioned, the support for this hypothesis so far falls solely on a somewhat incomplete theoretical model.
Single nucleotide polymorphisms (SNPs) exist in clusters of varying size and density across the genome. Despite this non-random distribution, scientists believed for many years that these so-called mutational hotspots were the product of natural selection and other post-mutational processes, and that the mutations occurred at random. However, "in last two decades, the large amount of both genomic and polymorphic data has changed the way of thinking in the field," Dacheng Tian of linkurl:Nanjing University;http://www.nju.edu.cn/cps/site/njueweb/fg/index.php in China, who did not participate in the work, wrote in an email to The Scientist
. "[This] idea provides a self-increasing hypothesis, which may be useful to rethink the formation of such non-randomness."
In the Royal Society
study, evolutionary geneticist linkurl:William Amos;http://www.zoo.cam.ac.uk/zoostaff/amos.htm of Cambridge University suggested one of the first mechanisms by which mutations may occur non-randomly. In this theoretical paper, Amos proposed that SNPs may form more commonly around pre-existing mutations as a result of a DNA repair system. When the repair system encounters a mismatch in the genome, which may occur at heterozygous regions where SNPs already exist, the repair machinery rips up the mistake and relays the DNA to correct it. Because mutations can occur when DNA replicates, the extra rounds of DNA replication associated with the DNA repair system could potentially cause mutations to cluster.
Using SNP data from the HapMap website for human chromosome 1, Amos calculated the average size and density of existing mutational clusters. He then ran simulations under the assumption of either this new, non-random mechanism of mutation formation or that of randomly occurring mutations. He found that the non-random model more closely predicted the frequency and density of the mutational clusters on the chromosome.
"My theory is going to shake things up majorly," Amos said. "The concept of non-independent mutations simply wasn't thought of before -- this is completely new and it really changes how we think of DNA evolving."
One interesting implication of this mechanism of SNP formation is that "it attracts mutations to where polymorphisms already exists, where it is likely to be tolerated [or even] beneficial," and vice versa, Amos said. "If you bias the mutations that do occur to where other mutations [already exist], you're more likely to do good than" if mutations occurred randomly. This mechanism, Amos added, may thus provide a way for the genome to reduce the overall number of deleterious mutations that occur.
While the idea is interesting and "it may be true," Tenaillon said, "I'm not convinced." There are several factors known to contribute to the non-random distribution of SNPs that Amos did not include in his simplified model, such as natural selection and demography. The HapMap data Amos used, for example, come from a European population, which is widely believed to have undergone a major bottleneck about 30,000 years ago, Tenaillon said. Such a bottleneck can significantly alter the SNP distribution, causing enormous increases in the numbers of SNPs in certain areas of the genome, she explained. Furthermore, this data set included both coding and noncoding regions, which are known to vary in the density of mutations since natural selection acts more potently on coding regions.
"It's not easy to discriminate between all these mechanisms," Tenaillon said. "It would have been nice if they could have taken into account all the things we know can create mutational hotspots and show that [the] effect [of self-perpetuating SNP formation] was [still] significant." More rigorously testing this hypothesis could include creating a more comprehensive theoretical model, as well as using genomic data from noncoding regions only and from a more "worldwide" population, she said.
Still, the idea warrants further exploration, Tenaillon added. "It's nice to have a paper where you have an idea it gives you material to discuss something," she said. "It's an interesting [concept] to test."
Clarification: In the original version of this story, a quote from Amos could have been misinterpreted to mean that mutations occurring under the proposed nonrandom mechanism were more likely to be beneficial than deleterious. Rather, the chance of being beneficial will be higher under this mechanism than if mutations occurred randomly.
**__Related stories:__***linkurl:Phase I of HapMap Complete;http://www.the-scientist.com/news/20051026/01/
[26th October 2005]*linkurl:Whole-Genome SNP Genotyping;http://www.the-scientist.com/article/display/13828/
[2nd June 2003]*linkurl:SNPs as Windows on Evolution;http://www.the-scientist.com/article/display/12778/
[7th January 2002]