Haplogroup R1b | Team USA

Admin
Administrator

Posts: 82,201

Haplogroup R1b Jan 14, 2021 4:59:18 GMT

Quote

Post by Admin on Jan 14, 2021 4:59:18 GMT

To increase the confidence level, the correlation of the
growth rate of cases and deaths with R1b percentages is
assessed also for the extended sample B. The results exceed
5 significance (Table 2).

It is apparent that all countries with high R1b percentage
have had a high growth rate of the contagion (Figure 4).
In countries with low R1b percentage the influence of the
other haplogroups is not negligible, therefore the effects are
less clear-cut. Besides, data for countries with low α or low
R1b percentage may be proportionally more noisy. Data for
low GDP countries could also correlate less well because of
lack of testing or late testing. However the situation of the
outliers should be examined on a case-by-case basis.

In fact, this correlation works particularly well in Western
Europe, where haplogroup R1b is the most frequently occurring
paternal lineage. One possible reason could be that a
viral strain which is more contagious for individuals belonging
to this haplogroup has got selected. The strain carrying
the mutation D614G, which began spreading in Europe in
early February, would probably be a good candidate [23].

Another problem is that the method used to calculate
the rates α may be slightly biased. By the time countries
with small populations reach 30 cases, they may have already
enforced containment measures, which could result
in underestimated rates α. This can also be verified performing
a simple linear regression with rates α and population
size, which gives significant results (on sample B
R = 0:33561, p-value = 0:001803 for cases and R = 0:27795,
p-value = 0:01047 for deaths). To eliminate this confounding
factor, one solution could be to choose the starting day of
the data set using, as a reference point, a number of cases
(or deaths) proportional to the population size. Another
possibility is to use the maximum value of α computed varying
the starting day over the entire infection period. These
solutions could be developed in future works.

Last Edit: Jan 14, 2021 5:02:36 GMT by Admin

Admin
Administrator

Posts: 82,201

Haplogroup R1b Jan 14, 2021 21:29:39 GMT

Quote

Post by Admin on Jan 14, 2021 21:29:39 GMT

5 Correlation with cases and deaths
per capita
A somehow complementary approach could be to consider a
later period, when all the countries have already enforced
containment measures. At this time, it seems more appropriate
to just compare the haplogroup percentages with the
number of cases per capita for each country, like we did
visually at the beginning. The correlation is assessed on
sample A in order to have a set of countries with comparable
starting times of the epidemic. For example, using
2020 world population estimates [24] [25], we can assess the
correlation with the number of cases on the day Figures 1
and 2 refer to (Table 3a).

We find again a highly significant correlation with haplogroup
R1b and no significant association with the other
ones. It is interesting to study how the value of the correlation
with average and maximum R1b percentages varies
over time (Figure 5a).

In the first weeks, the values are not significant because of
the limited diffusion of the virus across the countries in the
sample. After reaching a peak, they decrease, probably as an
effect of the different containment policies adopted among
countries. Moreover, the correlation with the maximum
percentages is stronger than the correlation with the average
ones at the beginning, and weaker afterwards. Such behavior
could be explained with the hypothesis that the effects of
contagion become evident first in areas with higher R1b
frequency. Once the virus has spread more evenly across the
countries, the effects of the average R1b frequency are more
relevant.

To avoid the delay due to the cumulative count of cases, we
can study the correlation with the daily number of cases per
capita, averaged on a week to reduce the weekly fluctuations
(Figure 5b). First the maximum percentages prevail over the
average ones, then the situation get reversed and lastly the
maximum percentages prevail again. Lockdowns, severing
the connections between people, create a situation arguably
more similar to an even distribution of R1b carriers. As a
consequence, this behavior could also be related to the onset
of lockdowns and to their lifting. Notably, a very strong
correlation with average percentages results on April 10th
(R = 0:85909, p-value = 2:4514 10􀀀17).

In the last observed days the correlation with R1b becomes
negative. The reason could be that the countries with higher
R1b percentages were struck earlier than the others. Now
they are in the descending phase of the contagion. On the
contrary, in countries with lower R1b percentages where the
virus has begun spreading later, the number of new cases is
still growing.

Last Edit: Jan 14, 2021 21:30:22 GMT by Admin

Admin
Administrator

Posts: 82,201

Haplogroup R1b Jan 15, 2021 7:34:04 GMT

Quote

Post by Admin on Jan 15, 2021 7:34:04 GMT

As before, we can apply the same approach to the number
of deaths per capita (see also [17]) and calculate the corre-
lation with haplogroup percentages, still on the same day
(Table 3b).

We notice once more a significant correlation only with
haplogroup R1b. In this case too, if we plot the correlation
with average and maximum percentages of haplogroup R1b
over time (Figure 5c), the difference between them tends to
progressively decrease. Nevertheless, the correlation with
the maximum percentages is always stronger than with the
average ones. This is probably due to the delay of a couple
of weeks between case detection and decease and to the
decreasing case fatality rate [26], which makes the total
death count change proportionally slower than the total case
count. The latter statement is confirmed by the study of
the correlation with the daily number of deaths per capita,
still averaged on a week (Figure 5d). It is apparent from
the graph that in the last observed period the correlation
is not significant anymore (the two fluctuations are due to
Spain death recounts on May 25th [27] and June 19th [28]).

Admin
Administrator

Posts: 82,201

Haplogroup R1b Jan 15, 2021 20:56:13 GMT

Quote

Post by Admin on Jan 15, 2021 20:56:13 GMT

6 Possible interpretations
Of course correlation does not imply causation and indeed
there are many environmental factors that can help the
virus to spread and there could exist environmental factors
that favored the haplogroup R1b diffusion (e.g. like malaria
selected the ‘favism gene’ [29]). But it seems unlikely there
could be one environmental factor that does both, because
of the very different timescales it should operate on (R1b
emergence dates back to at least 14,000 years ago [30]).
Assuming the existence of an underlying genetic factor,
one possibility would be that the interested genes are located
on the sex chromosomes, considering the higher lethality
for men [31]. They could be directly on the Y chromosome,
being R1b a Y-DNA haplogroup, or even on the X chromosome
(‘unguarded’, like in the case of color blindness [32]),
especially since the ACE2 gene is located on the X chromosome
and the human ACE2 receptor is known to be involved
in the transmission of the virus [33].

Indeed, the difference in lethality between sexes is correlated
with the haplogroup R1b frequency in the population,
but this probably happens because they are both correlated
with the case fatality rate [15]. One alternative explanation
for the Y chromosome signal is that we are tracing an
autosomal locus associated with the haplogroup R1b, like
the ones identified in recent studies [34] [35] [36]. Unless
differently specified, we will loosely refer also to those people
as ‘R1b carriers’.

Three scenarios, not necessarily mutually exclusive, seem
to be compatible with our findings (Table 4 ). An R1b carrier
could even experience all of them, at different stages of the
disease.

In the first scenario, infected R1b carriers tend to develop
very severe symptoms and therefore are more likely to be
detected, resulting in a higher number of cases where the
haplogroup frequency is higher. In this case, people belonging
to the other haplogroups could be infected as well, with
few or no symptoms, making them harder to discover.

In the second scenario, R1b carriers are more susceptible
to infection. It could be even hypothesized that the more
an individual is susceptible to infection, the higher the risk
of dying if infected, which seems to hold true at least for
children and old people [37] [38]. This could be another
reason why the correlation of R1b with deaths is stronger
than with cases. It could also help to explain why in Italy,
initially, a very high case fatality rate was faced [39], but
lately the virus seems to have lost strength [40], considering
that the Italian population is genetically stratified. After the
lockdown, with the virus less widespread and having been
the more susceptible individuals already infected, it is less
likely for the virus to encounter new clusters of susceptible
individuals.

Probably in a similar way, the virus was already spreading
earlier in Italy [41], since December according to [42],
undetected. In this initial stage, the virus could have gone
unnoticed because it spread among less susceptible individuals
at a slow pace, because of the low (effective) reproduction
number, and probably killing few people. Only when it encountered
a susceptible community, the infection quickly
grew and got noticed. Since the initial spreading rate correlates
better with maximum R1b percentages than with
average ones, it can be supposed that similar dynamics were
at work also in other countries.

In the third scenario, R1b carriers are more contagious,
being less symptomatic, having higher viral shedding, or
both. Indeed, asymptomatic carriers [43] [44] and superspreaders
[45] seem to play an important role in COVID-19
pandemic. A higher viral shedding may produce in the exposed
individuals a higher viral load and, therefore, more
severe infections [46]. This could be another reason why the
correlation of R1b with deaths is stronger than with cases.

Admin
Administrator

Posts: 82,201

Haplogroup R1b Jan 15, 2021 23:38:32 GMT

Quote

Post by Admin on Jan 15, 2021 23:38:32 GMT

In areas where there is a higher R1b frequency, they could
spread the contagion, affecting people belonging to other
haplogroups. This could be one of the reasons why black
people and other ethnic minorities are dying in disproportionate
amounts in UK [47] and US [48], where they live
intermixed with R1b carriers, but not in Africa or Asia. It
could also be one of the reasons why the correlation with
R1b percentage becomes negative in the last period of observation.
Asymptomatic carriers are less likely to be counted
as cases and, once social distancing measures are enforced,
they do not spread the virus either.

Moreover, because of insufficient data, we cannot exclude
a prevalence of males among asymptomatic carriers.
Thus, the involved genes could even be located on a sex
chromosome. In particular, the Y chromosome can regulate
the expression of various genes of the autosomes and
X chromosomes [49], and males belonging to haplogroup
R1b possibly have more effective immune and inflammatory
responses [50]. This could make infected R1b carriers
more likely to be asymptomatic, or paucisymptomatic. Besides,
there is a significant negative correlation between
the percentage of males over the total cases and the R1b
percentage (R = 􀀀0:40884, p-value = 0:0058619 on sample
C; R = 􀀀0:38441, p-value = 0:0017122 on sample D).
Considering the positive correlation with the number of
cases per capita, this appears to imply that R1b carriers are
asymptomatic and, therefore, more difficult to detect.
In order to validate one or more of the aforementioned
scenarios, a study of the genetic profile of the population in
the most affected areas would be needed, such as the study
in Vo’ (Veneto) [4].

7 Conclusion
In this work, first, a coefficient α representing the growth
rate of the contagion is obtained for 56 countries, using an
exponential fit of the number of cases in the first days of
spreading. Then, the Pearson’s linear correlation coefficient
between the rates α and the average Y-DNA haplogroup
percentages in the corresponding countries is calculated, obtaining
a highly significant correlation with haplogroup R1b.

Lastly, the correlation between the same rates α and the
maximum haplogroup R1b percentages in the corresponding
countries is calculated, obtaining an even more highly significant
correlation. A similar calculation is repeated with the
number of deaths in the initial days of spreading, obtaining
again highly significant correlation with haplogroup R1b.
The same procedure is then repeated with an extended R1b
data set of 84 countries, strengthening the results.
Moreover, it has been studied how the correlation of average
and maximum haplogroup R1b percentages with cases
per capita varies over a five-month period. The same procedure
has been applied to the deaths per capita. Comparing
the results, it emerges that at the beginning the correlation is
stronger with the maximum R1b percentages than with the
average ones, but over time the difference decreases. This
could be explained supposing that the effects of contagion
become evident first in areas with higher R1b frequency.
Three possible scenarios are then outlined, according to
the passive or active involvement of R1b carriers in the diffusion
of the virus. They could develop more severe symptoms,
be more susceptible to infection, be more contagious, or one
of their combinations. In particular, R1b carriers being
asymptomatic (or pre-symptomatic) spreaders seems to be
the most likely option.

Given those results, it could be useful to take into consideration
the ideas here presented when confronting the current
pandemic struggle. Our findings should warrant further
epidemiological observational studies, such as case-control
studies, to confirm or disprove the possible involvement of
R1b carriers in the spread of the virus. A positive result
could speed up the discovery of a treatment, help to make
more reliable quantitative forecasting models or, at least,
help to better tune the social distancing measures and to
inform future vaccination campaign priorities.