The Yamnaya Steppe Herders from Russia

new

Admin
Administrator

Posts: 80,366

The Yamnaya Steppe Herders from Russia Mar 7, 2021 22:24:46 GMT

Quote

Post by Admin on Mar 7, 2021 22:24:46 GMT

A potential limitation of our approach, which is
inherent to most aDNA studies, is genetic discontinuity
due to large population replacements or to sampling bias
for geographical locations.35 For example, different sampling
proportions from northern and southern Europeans
across epochs may result in genetic discontinuity in our
dataset, given that the former present higher Eastern
steppe ancestry than the latter after the Bronze Age.36 We
thus repeated our ABC setup for northern and southern Europeans
using a geographical division,37 designed to distinguish
high and low levels of Steppe ancestry (Figure S8).

Despite much lower sample sizes, we found evidence for
negative selection in both northern (s ¼ 0.24; 95% CI:
[0.02–0.87]) and southern (s ¼ 0.13; 95% CI: [1.6 3 10
4–0.81]) European homozygotes, with a slightly left-shifted
posterior distribution in southern Europe, where the sample
size is more limited (Figure S9). We also found, using
factor analysis,38 that P1104A carriers scattered
throughout European sub-structured populations, across
all epochs after its introduction to Europe (Figure 4).



In addition, ancestry proportions were similar between
P1104A carriers and the rest of the dataset at each epoch (Table
S1). Notably, the observed ancestry shift between Bronze
Age and present-day samples (from 0.29 to 0.36 for the whole
dataset [Table S3], representing a 24% relative increase, and
from 0.23 to 0.39 for P1104A carriers [Table S1]) does not,
on its own, explain the frequency decline of the allele after
the Bronze Age (from 0.074 to 0.029, representing a 61% relative
decrease). Yet, we performed an ABC estimation accounting
for ancestry variation across epochs (supplemental
material and methods). Using the estimated Anatolian
ancestry of our dataset at each epoch fromthe Late Neolithic
onward, we estimated very similar values for the strength
and onset of negative selection for TYK2 P1104A at the
pan-European level (s ¼ 0.27; 95% CI: [0.08–0.93]; Tonset ¼
2,045 ya; 95%CI [500–8,690]; Figures S10A and S10B). Similarly,
we found comparable estimations for northern and
southern Europeans (s ¼ 0.26; 95% CI [0.06–0.83]; Tonset ¼
1,046 ya; 95% CI [500–6,934]; Figures S10C and S10D; and
s ¼ 0.24; 95% CI [0.02–0.85]; Tonset ¼ 3,229 ya; 95% CI
[500–8,963]; Figures S10E and S10F, respectively).
Conversely, we found no evidence of selection for TYK2
I684S (s ¼ 0.02; 95%CI: [0–0.69]), as expected, and a weaker
signal of negative selection for HFE C282Y (s¼0.12; 95%CI:
[0–0.76]). Collectively, these findings suggest that the
observed frequency drop of P1104A after the Bronze Age is
not due to major geographical and/or temporal differences
in ancestry components in our aDNA dataset, but instead
to the action of natural selection. Moreover, when re-estimating
the age of P1104A without modern data from Middle
Easterners and Central Asians, as they are not entirely representative
of ancestral Anatolian farmers and steppe herders,
respectively,39,40 we obtained almost identical results (mode
¼ 30,303 ya; 95% CI [23,113–60,273]).

Last Edit: Mar 7, 2021 22:30:24 GMT by Admin

Admin
Administrator

Posts: 80,366

The Yamnaya Steppe Herders from Russia Mar 8, 2021 4:12:41 GMT

Quote

Post by Admin on Mar 8, 2021 4:12:41 GMT

In this attempt to define the mode and tempo of TB pressure
on Europeans, we found that the only common variant
known to underlie monogenic predisposition to TB has
evolved under strong negative selection in Europe after
the Iron Age. In doing so, we provide population genetic evidence
for the high burden of TB in Europeans over the last
twomillennia, in line with the dating ofM. tuberculosis lineage
4 at 1,943 ya41 and of strains found in 18th century Hungarian
mummies at 1,604 ya, or in mummified remains of
the 17th century Bishop Peder Winstrup of Lund between
929 and 2,084 ya.19,31 Notably, the TB-associated mutation
ranks among the top 2.7% of variants, present in the studied
capture array, with similar frequencies in the Bronze Age
(0.04–0.10) that have decreased the most since this period
(Table S3; supplemental material and methods). Such variants
might also include targets of negative selection (Table
S5). A selection coefficient of 0.20 would entail
>2,500,000 cumulative deaths over the last 2,000 years
due to P1104A homozygosity, representing 1%–2% of all
TB-related deaths in the 19th century Europe (Figure S11).
This figure is consistent with a previous estimation of 1%
of TB cases due to the at-risk genotype among present-day
Europeans.7 We anticipate that the same population genetics
framework could be used to delineate other human
genetic variants, of yet unknown function, that have drastically
decreased or increased in frequency across time due
to microbial pressure. Thus, adopting an evolutionary
approach represents a promising alternative to investigate
the genetic sources of present-day disparities, between individuals
and populations, in susceptibility to infection.

Data and code availability
Pseudo-haploid ancient and modern genome data are
available at reich.hms.harvard.edu/allen-ancientdna-
resource-aadr-downloadable-genotypes-present-dayand-
ancient-dna-data (V42.4: March 1, 2020 release). Code
to perform ABC estimations from simulated frequency
data are available at github.com/h-e-g/SLiM_
aDNA_selection.

Supplemental Information
Supplemental Data can be found online at doi.org/10.
1016/j.ajhg.2021.02.009.

Admin
Administrator

Posts: 80,366

The Yamnaya Steppe Herders from Russia Mar 10, 2021 22:12:10 GMT

Quote

Post by Admin on Mar 10, 2021 22:12:10 GMT

Heterogeneous Hunter-Gatherer and Steppe-Related Ancestries in Late Neolithic
and Bell Beaker Genomes from Present-Day France

SUMMARY
The transition from the Late Neolithic to the Bronze Age has witnessed important population and societal changes in western Europe.1 These include massive genomic contributions of pastoralist herders originating from the Pontic-Caspian steppes2,3 into local populations, resulting from complex interactions between collapsing hunter-gatherers and expanding farmers of Anatolian ancestry.4–8 This transition is documented through extensive ancient genomic data from present-day Britain,9,10 Ireland,11,12 Iberia,13 Mediterranean islands,14,15 and Germany.8 It remains, however, largely overlooked in France, where most focus has been on the Middle Neolithic (n = 63),8,9,16 with the exception of one Late Neolithic genome sequenced at 0.053 coverage.16 This leaves the key transitional period covering 3,400–2,700 cal. years (calibrated years) BCE genetically unsampled and thus the exact time frame of hunter-gatherer persistence and arrival of steppe migrations unknown. To remediate this, we sequenced 24 ancient human genomes from France spanning 3,400–1,600 cal. years BCE. This reveals Late Neolithic populations that are genetically diverse and include individuals with dark skin, hair, and eyes. We detect heterogeneous hunter-gatherer ancestries within Late Neolithic communities, reaching up to 63.3% in some individuals, and variable genetic contributions of steppe herders in Bell Beaker populations. We provide an estimate as late as 3,800 years BCE for the admixture between Neolithic and Mesolithic populations and as early as 2,650 years BCE for the arrival of steppe-related ancestry. The genomic heterogeneity characterized underlines the complex history of human interactions even at the local scale.

Figure 1. Sample Location and Radiocarbon Dates
(A) Geographic location of Mont-Aime hypogea. The city of Reims, located north of Mont-Aime, is shown.
(B) Geographic location of the main regions of present-day France investigated in this study. The city of Narbonne, located near the La Clape massif, is shown.
(C) Zoom in of the archaeological sites from southwestern France. The number of individuals analyzed genetically is indicated (n = ).
(D) Bayesian posterior distributions of radiocarbon dates calibrated with OxCal online (https://c14.arch.ox.ac.uk/oxcal/OxCal.html) on IntCal20.18
See also Figure S1 and Data S1.

RESULTS AND DISCUSSION
Archaeological Sites and Radiocarbon Dates
A total of 30 ancient human petrosal bones, 5 post-cranial bones,
and 18 teeth were prepared in ancient DNA facilities for shotgun
sequencing on Illumina platforms (STAR Methods). Following
shallow sequencing on the MiniSeq instrument, 26 (49%)
DNA extracts showing at least 10% of human DNA were selected
for further sequencing on the NovaSeq and HiSeqX instruments
(Data S1A). These correspond to remains excavated from
collective burials associated with Late Neolithic, Bell Beaker, and
Early Bronze Age archaeological contexts and located in the
Paris Basin (Mont-Aime) and southwestern France (Mas Rouge,
Grotte du Rouquet, Grotte Basse de la Vigne Perdue, and Grotte
des Tortues; Figures 1A–1C; STAR Methods). Direct radiocarbon
dating indicated that the two Mont-Aime hypogea (i.e., humanmade
subterranean burials) were active between 3,400 and
2,900 cal. years (calibrated years) BCE (Figure 1D). All six individuals
from Grotte du Rouquet were virtually contemporary to
those from Mont-Aime. Grotte des Tortues included one Late
Neolithic individual dated to 3,100–2,900 cal. years BCE
(TORTD), and three individuals spanning the mid-3rd (TORTC
and TORTE) and the first half of the 2nd (TORTF) millennium
BCE. This indicates that the Grotte des Tortues cave was used
opportunistically as a collective burial site over a time period encompassing
the Late Neolithic to the Early Bronze Age, in line
with previous archaeological evidence.17 The same was true
for Grotte Basse de la Vigne Perdue, although only from the
mid-3rd to the mid-2nd millennium BCE.

Last Edit: Mar 10, 2021 22:14:10 GMT by Admin

Admin
Administrator

Posts: 80,366

The Yamnaya Steppe Herders from Russia Mar 11, 2021 6:00:13 GMT

Quote

Post by Admin on Mar 11, 2021 6:00:13 GMT

Genomic and Metagenomic Content
Metagenomic taxonomic profiles inferred frommetaBIT19 and the
MetaPhLAn2 database20 mainly contained contaminating bacterial
species of unknown or environmental origin (Data S1B). Three
out of 10 teeth fromMont-Aime (2H10, 2H11, and 1H13), however,
showed increasing fractions of human oral bacteria (4.8%–
39.5%). These include the cariogenic Streptococcus mutans, as
well as Treponema denticola, Porphyromonas endodontalis, and
Campylobacter rectus that are associated with peri- and endodontal
diseases21 but were also detected in healthy individuals
today and in 120- to 5,700-year-old samples worldwide (Data
S1D).22–34 The presence of S. mutans as early as the Neolithic
adds to the evidence27,34,35 that this pathogen cannot provide a
reliable marker for tooth decay following lifestyle changes in the
Bronze Age and post-industrial era, despite earlier claims.36 The
proximity of oral metagenomic profiles measured in 1H13 and
modern individuals (Figure 2A) is in line with the reduced fraction
of unclassified and environmental microbes in this sample but
may also indicate thatmicroscopic calculus remained undetected
and was co-extracted with the tooth cementum.
Mitochondrial sequence alignments indicated minimal human
DNA contamination in all individuals (Data S1E), except two from
Mont-Aime; 2H06, which is characterized with minimal
sequence data (and excluded from downstream analyses), and
1H14, which shows 10%–12% contamination. X-to-autosomal
sequence coverage revealed 13 males, for whom X chromosome
heterozygosity profiles confirmed minimal autosomal
contamination (%1.01%; median = 0.12%; Data S1F).

Figure 2. Microbial Profiling, DNA Methylation Analyses, and Phenotype Prediction
(A) Genus-level principal-coordinates analysis (PCoA) of bacterial profiles. Taxa supported by abundances strictly inferior to 1% were disregarded.
(B) Correlation matrix of DNA methylation levels at gene bodies.
(C) Genotypes observed for the two Mont-Aime samples sequenced to high coverage at 49 loci underlying important morphological, metabolic, behavioral, and
medical phenotypes. The number of independent sequence counts is provided for each locus considered. Undetermined genotypes are indicated as ‘‘./.’’
See also Data S2 and S3.

Admin
Administrator

Posts: 80,366

The Yamnaya Steppe Herders from Russia Mar 11, 2021 21:20:50 GMT

Quote

Post by Admin on Mar 11, 2021 21:20:50 GMT

Genetically identified as female, individual 1H14 could thus not
be tested with this approach. Sequence data displayed post
mortem DNA degradation characteristic of DNA extracts partially
treated with USER37 (Figure S1A), and genetic affinities were in
line with those Mont-Aime males showing minimal contamination.
As mtDNA data do not necessarily reflect autosomal
contamination rates,38 individual 1H14 was included in downstream
analyses. Overall, we report a total of 24 ancient genomes
at 0.20- to 23.88-fold average depth of coverage (median
= 0.96-fold), including two at high depth (2H10 and 2H11,
13.91- and 23.88-fold, respectively). Supporting our data quality,
error rates ranged between 0.057% and 0.083% substitutions
per base (median = 0.069%), which is approximately half the
average error rate estimated in ancient (high-coverage) genomes
previously reported (Figures S1C–S1E).

Uniparental Markers and Kinship
No individuals from the same archaeological site carried identical
mitochondrial haplotypes, indicating no maternal relatedness
(Data S1E). In contrast, all males were assigned to the Y
chromosome haplogroup I2a1, except one individual at Mont-
Aime belonging to haplogroup H2a1 and one 4,400-year-old
R1b1a1b1a1a2a1 individual from Grotte Basse de la Vigne
Perdue (Data S1F). Close or incompletely covered but matching
haplotypes were previously reported in contemporary Bell
Beaker individuals (R1b1a1b for sample CBV95, La-Bouche-a` -
Vesle) and Bronze Age individuals from both northern (R1b1a1-
b1a1a2 for sample RIX2, Rixheim-Zac du Petit-Prince; R1b1a1-
b1a1a2a5 for sample OBE3626-1, Obernai PAEI) and southern
France (R1b1a1b1a1a for sample PIR3116B Rec de Ligno).16
This may reflect admixture from incoming steppe herders that,
around that same time, have almost completely replaced other
Y chromosome haplotypes in Iberian populations.13 Haplogroup
I2a1 was previously found among various hunter-gatherer
groups from Europe.4,6,12,13,39,40,41 This may indicate the persistence
of western hunter-gatherer (WHG) ancestry in at least
some Mont-Aime individuals (2H11).

The apparently high frequency of I2a1 may reflect a general
dominance of this haplogroup across Late Neolithic France. It
may, however, also suggest communities organized patrilinearly,
especially as individual sites showed no mitochondrial
haplotypes shared between individuals. To explore this further,
we assessed patterns of genetic relatedness at 1,240k autosomal
SNP positions40 using pseudo-diploid calls in READ42
as well as genotype likelihood in lcMLkin43 (Data S1G and
S1H). This indicated that five individuals analyzed at Mont-
Aime were genetically related. In hypogeum 2, a father and his
son (2H10 and 2H17) were related to a lesser degree to a third
male (2HxC5x196x1131). They shared similar Y chromosomal
haplotypes, confirming patrilinearity. However, we also identified
a father-daughter relationship between two individuals anatomically
determined as adults (2H11 and 1H06; Data S1A). That at
least some women could remain in their native communities implies
that no strict patrilocal rules were followed. Alternatively,
this could also imply that at least a fraction of women were not
involved in marital relationships or had their bodies repatriated
to their father’s community after death. Extending the analyses
to more individuals and integrating isotopic signatures informing
on birth location will help assess the extent to which this also
applied to other women in the community. Nonetheless, the
presence of first-degree relationships between Mont-Aime I
and Mont-Aime II supports that both hypogea were used strictly
contemporaneously and by the same group of people.

Inbreeding and Population Diversity
The fraction of the genome that is comprised within runs of homozygosity
(ROHs) can indicate within-kin unions.12 It can also
result in under-estimating genome-wide levels of heterozygosity,
which inform on the population effective size.44 In order to both
assess the presence of within-kin unions and estimate population
effective sizes at Mont-Aime, we applied ROHan44 to the
two individuals sequenced to high depth (2H10 and 2H11). The
ROH genome fraction was, however, limited (1.5%–2.0%; Data
S2A) and on par with that observed among other Neolithic individuals,
except at the megalithic site of Newgrange, where a
brother-sister incestuous union was previously reported.12 This
indicates limited, if any, within-kin unions at Mont-Aime. Estimated
heterozygosities outside ROHs were comparable to those
found in other high-coverage Neolithic genomes, except Newgrange,
but larger than those of west European Mesolithic hunter-
gatherers (Loschbour and Sramore62; Data S2A). This is in
line with Neolithic food production economies allowing population
demographic expansion relative to hunting-gathering
communities.45