Reversions, also known as back mutations are considered rare in biology. Basically, what it means is an initial mutation in the geneome, reverses back to its original (wild-type) form through a second mutation that restores the base in the DNA sequence.
For example, a "chunk" of DNA could contain the following bases: ACGCTG and, a random, chance mutation replaces the cytosine (C) for an adenine (A) ACGATG and a second mutation restores (reverses) the original situation ACGCTG.
This has important consequences, first of all, if we look at the ancient sample and the most recent one, there is no way we will ever know that it mutated and reverse by back-mutating, (both have the same sequence: ACGATG so how can we know if there was a back and forth flip in between both samples?)... unless we find an sample of someone in the same line, or a parallel lineage with the first (derived) mutation, but not the reversion,which seems a very unlikely situation.
Assuming that all mutations are forward oriented and never reverse, may overlook mutations that were reversed. Since coalescence time and dates of lineage splits are based on mutations, if we overlook the reversions, we will miss out on the actual mutations (to and fro), counting zero when in fact there were two mutations.
So we will assume that mutation rates are lower than they really are by missing out these back-and-forth mutations.
If we overlook reversions we will assume there were only n mutations per a given amount of years, while there were actually m mutations: "n" that we see (for instance, there is a G instead of an A at a certain locus), and "p" mutations that flipped forth and another "p" that flipped back. n is therefore smaller than m; m = n+2p. So the mutation rate is higher than assumed.
However, the Neutral Theory of genetic evolution does not consider this alternative, it requires No back-mutations. Changes can only happen in one direction A → G. Which will never again flip back G → A, and No Recurrence there can't be multiple mutations at identical loci in different lineages.
Are they Common?
William Amos (2020) suggests that "back-mutations are far commoner than has been previously assumed" he adds that "Back-mutations are ‘silent' because they create the original ancestral allele, but can reasonably be assumed to occur about twice as often as triallelic SNPs are generated (two transitions are approximately twice as likely as one transition and one transversion). Triallelic SNPs are coded ‘MULTI-ALLELIC' rather than ‘SNP' in the 1000 genome data and are often ignored, but I counted 257,827 occurrences across all autosomes, implying over half a million sites carrying back-mutations. Moreover, this is probably an underestimate because the 1000 genomes data are low coverage and rely on extensive imputation which will often cause rare third alleles to go undetected. Equally, conservative curation will tend to remove third alleles that lack strong support. Note, triallelic sites are unlikely to be generated mainly by sequencing errors because only 1% of these sites carry a singleton as the rarest allele. This analysis is not intended to provide an accurate estimate of the back-mutation rate, but instead simply to demonstrate that large numbers of back-mutations do exist to the extent that models of evolution that rely on back-mutations occurring in appreciable numbers should not be dismissed a priori."
Research by Anke Fähnrich et al., (2023) on the North and Eastern African mtDNA shows that some mutations that serve as markers appear time and time again, the authors consider some as "Shared back mutations", others are simply repeat mutations. The paper shows them in a tree for "L0a1 and (b) L2a1" and clarifies that "We highlight with magenta, gray and turquoise boxes those variants that indicate that a different phylogenetic tree may better explain the samples from North and East Africa." These trees can be seen in the image below. The paper adds that "Shared back mutations (magenta) denote that parental haplotypes may be missing in PhyloTree. Mutations repeatedly observed in a subtree (gray) suggest that child haplogroups are missing. Variants that occur in multiple samples and differ from variants defining a parental haplogroup (turquoise) suggest that different variant combinations and haplogroup specifications may better explain North and East African mtDNA sequences". It also marks them with an "@" as "assumed back mutation or missing mutation". This goes to show that markers are shared across different haplogroups.
A similar situation was reported by Neil Howell, Joanna L Elson, D M Turnbull, and Corinna Herrnstadt (2004) who were investigating the oldest mtDNA haplogroups L0 and L2. Besides finding oddities in the trees, ages, etc., they noted multiple reversions: "The L0a outgroup sequence carries C alleles at nucleotides 16189 and 16192, whereas the L2a ancestral sequence is predicted to carry C and T, respectively, at these sites. The 16189 site subsequently undergoes mutation on four occasions (three forward and one reverse relative to the outgroup sequence), whereas the 16192 site undergoes reversion on five occasions. Thus, both sites appear to have relatively high rates of mutation, a result that has been observed in previous studies (Excoffier and Yang 1999; Meyer, Weiss, and von Haeseler 1999; Howelland Bogolin Smejkal 2000) and in the L2a networks of Salas et al. (2002)... The ancestral L2a sequence carries a C:T transitional nucleotide 16519, which undergoes reversion on three occasions. These results are not surprising and this site has long been recognized to have a high mutation rate."
This seems to define a "mutation hotspot", that I mentioned in a previous post (Laguna de los Pampas 10,000 BP remains in Argentina).
Patagonian Monsters - Cryptozoology, Myths & legends in Patagonia Copyright 2009-2026 by Austin Whittall ©
No comments:
Post a Comment