“Mitochondrial DNA and Human Evolution” was the title of a scientific article that appeared in the January 1, 1987 issue of
Nature,
authored by Rebecca Cann, Mark Stoneking, and Allan C. Wilson. These three scientists announced that all modern human beings can trace
their ancestry back to a single woman (the so-called
"Mitochondrial Eve"), who lived about 190.000 to 200.000 years ago in
Africa.
Scientists estimated that
"Y-Chromosomal Adam", the most recent common ancestor of men, lived much more recently, between 50,000 to 115,000 years ago.
The disparity between our most recent common ancestors (MRCA) might have resulted from limitations in gene sequencing: up until about five years ago, researchers could sequence only a few regions of the genome.
According to a couple of papers published on
Science
magazine the
2 August 2013 there is a good chance that Adam and Eve may have existed about the same time, evolutionarily speaking. Using the complete strand of DNA that determines male sex, researchers have determined that Y-Chromosome Adam lived 120.000 to 156.000 years ago, overlapping with Mitochondrial Eve, who lived 99.000 to 148.000 years ago.
Rebecca L. Cann (University of Hawaii at Manoa, Department of Cell and Molecular Biology) in her perspective for Science magazine (
Y Weigh In Again on Modern Humans) wrote: «So now it seems that a population giving rise to the strictly maternal and strictly paternal portions of our genomes could have produced individuals who found each other in the same space and tim
e».
Analyzing the Y chromosome sequences of
69 men from 9 populations, scientists from Stanford University and their colleagues have found that
Y-Chromosomal Adam did not live much more recently than
Mitochondrial Eve [Sequencing Y Chromosomes Resolves Discrepancy in Time to Common Ancestor of Males Versus Females].
Likewise, by
studying the genomes of
1.204 Sardinian men with a focus on the Y chromosome, a team based in Italy (University of Sassari, CNR, CRS4) calculated an age for the most recent common male ancestor that is consistent with previous estimates of the female ancestor based on mitochondrial DNA [
Low-Pass
DNA Sequencing of 1200 Sardinians Reconstructs European Y-Chromosome
Phylogeny].
Riccardo Berutti, one of authors of the second article, explains in this interview the importance of the Y chromosome analysis, that moves "Adam" closer to "Eve".
What you were most surprised of this study?
«In the last years along with the other studies we carried on within the
CRS4-CNR collaboration, we did a lot of work on the data for this
paper, and personally I learned a lot of stuff that time-by-time I
found surprising. There are three points that are worth to be
told. One of them is about the population dynamics in Sardinia.
You know that in our study we have a snapshot of the general
population. Although given the dynamics of an isolated population
like Sardinians it may appear as an obvious result, I found anyway
surprising that the whole paternal lineage of an important part of
the population of the I haplogroup, the one spotting private
Sardinian variability, never moved away from Sardinia since it
entered into the Neolithic, and so happened with other lines. We
arrived, we loved this island and never leaved. Fantastic, we are
still like that maybe. Secondly, I would point at the really
surprising precision we achieved on the individual samples, which
overtook by far the raw-data precision that we normally have on
sequencing data. We could reach such an accuracy by applying and
tuning appropriate filters both on raw data and on the obtained
genotypes and using the extremely powerful phylogenetic criteria,
which closed the games, leading almost to perfect results as compared
with other intrinsically more accurate genotyping methods. Just to
conclude this story, the third thing I really enjoyed is the extreme
versatility of genetic data. All Sardinian DNA we used in our study
belonged to two studies targeting to really different topics than
anthropology and phylogenetic stuff, just like the data we used as
outgroup (non Sardinians) that came mainly from the public archive of
the Thousand Genomes project. This to say that such data represent a
true gold mine for future studies and seen from this point of view,
the huge costs for the production of this data instantly turn into a
brilliant investment into the future of research».
What has surprised you most about this study?
«At least half of the samples where sequenced at CRS4, the other half at
the "Sequencing Core" facility of the University of
Michigan in Ann Arbor. With my colleagues at the AGCT (Advanced
Genomics Computing Technology) group we took raw sequencing data and
processed them up to obtaining aligned data. The alignment process
consists in rebuilding the complete genome for each individual sample
from short sequences which consist of 100 DNA bases each which should
be identified and matched to their place into the reference genome,
consisting of around 3 billion DNA bases. The same results were
collected from Ann Arbor. After the alignment step, data has a huge
error rate, in the range of 1-10% error, depending on the
version of the chemistry used, of the sequencing machine, and on
several other factors. We used sequences generated between 2008 and
2013 with huge differences in accuracy, while the latest have lower
error rates (even less than 1%) they are just a small part of the
whole cohort. All this means, one wrongly read DNA base every 100, or
even one every 10. And for every error you can spot a new mutation
which actually doesn't exist, or you can miss a mutation on the other
side. All that is quite normal with Illumina data and it is
fundamental, at this stage, to create and apply appropriate
filtering, depending on the dataset and on what you want to get from
it».
What was your role in this research?
«My personal role in all of that follows the raw data alignment, was firstly to quality
check every single sample, then to figure out which filters may suit
best to the Y chromosome and apply them. Such filters are usually
different from the ones that are commonly applied to the other
chromosomes, since Y is only carried in one single copy per male
sample, which means that you have less statistics with respect to the
other chromosomes, therefore you are more prone to errors. On the
other side you will never have heterozygous genotypes (i.e. when you
read at a certain position: A on the first copy of chromosome 1, and
C on the second copy, then you have an A/C heterozygous base call)
and this is a good way to filter. New software tools were necessary
and I took care of developing them, both for quality control and for
filtering on the Y chromosome. The downstream workflow was
approximately as follows: run a quality control test, apply quality
thresholds on the single samples, create statistical filtering on the
whole panel. And it's somehow a recursive process, you design your
filters, you see your results distribution and then tune the filter
chain. All this stuff was done continuously working side by side with
Paolo Francalacci, the first author, which worked on the phylogenetic
method and whose feedback was important for the amelioration of
filtering. It's worth to say that after all this work the precision
of the final data supplied to Paolo increased up to a maximum error
rate of 0.1%. After this processing Paolo applies the phylogenetic
criteria to build the trees and to improve even more the dataset, as
far as we could measure the results are nearly error-free. One
other part of my contribution was the analysis which allowed us to
infer the ancestral status of several alleles, that means, comparing
human mutations with chimpanzee ones on the Y chromosome. This was
particularly useful when establishing which mutations carried by the
A haplogroup, which is the African and the most ancient one, are
ancestral or A specific. In fact the reference genome has been built
from modern humans of more recent haplogroups than A, thus leading to
the fact that some of the reference genomes is mutated with respect
to the "original" homo sapiens. If you do not take into
account that you get inconsistent results».
Next step?
«We still have to go deep into the analysis of the mithocondrial DNA of the complementary of the source
datasets we used: the females. The so called mtDNA is small fragment
of genetic code, just 16.6 thousands bases long, really tiny compared
to the 8 million bases of the Y chromosome (over a length of 50
millions) that we used for our study. This fragment of DNA is the
'source code' for the mithocondria, which are a sort of power plant
for the cells. It's a good counterpart for the Y chromosome since it's maternally
inherited, so that it can tell us the other side of the story. We can
easily expect that while on the timing we can improve the previously
estimates, on the demographic history it's not straightforward that
we'll obtain the same tree as we did for males, and we could discover
a partially different path, that,once compared to the Y chromosome
history may improve our ability on doing population genetics using
the other chromosomes. Other steps will include sequencing ancient
bones from the various sites in Sardinia, such information might also
be helpful to better understand the specific Sardinian variability
leading to our peculiar phenotype and to the higher rate in Europe
for some auto-immune diseases. Some more plans are being made to
continue this collaboration with Francalacci and Cucca. They are
really two visionary men and have fantastic ideas about the future of
this research. In the meantime sequencing and computing technologies
are being continuously enhanced and lot of new things will become
possible. There's a really exciting path to dig into the secrets of
life.»
Riccardo Berutti, is a nuclear physicist with PhD in Genetics.
Andrea Mameli www.linguaggiomacchina.it (13 August 2013)
