Biology Reference
In-Depth Information
on the underlying genetic structure even when compared pairwise
between time periods for 18 households with sufficient sample sizes for
testing.
41
Furthermore, a spatial autocorrelation analysis showed that
parasites between households within 540 mwere more genetically similar
than expected by chance alone. Genetic differentiation measured as
F
CT
(hierarchical
F
-statistic of household to the total) was 0.023 and highly
significant (
p
0.0001)
41
.
These results
41
revealed three key insights into transmission of
A.
lumbricoides
in Jiri: there were separate foci of transmission at this local
scale, households and nearby houses shared genetically related para-
sites, and people reacquired their worms from the same source pool of
infection over time. These results challenge the dogma that a single
human community will correspond to a homogeneous parasite pop-
ulation (implicit in many classic models
44,45
of parasite transmission
that measure a single basic reproduction number,
R
0
). In Jiri, multiple
source pools of infection need to be considered when modeling para-
site transmission. Thus, when using models to evaluate control strate-
gies in Jiri, it would be more appropriate to consider incorporating
parasite populations that exist
<
in an interconnected network,
i.e.
metapopulation.
46
Although I emphasized how population genetics can be used to
elucidate transmission patterns, I note that I do not view landscape
genetics as a panacea for epidemiological goals in general, nor do I view
genetics data as a replacement for infection intensity data. Rather I see the
two types of data as providing different, but complementary, information
about the transmission process. For example, Walker and colleagues
32
found that in Bangladesh host age and sex explained part of the variation
in worm burdens. In contrast, host age and sex were not correlated to how
worm genetic variation was partitioned in Jiri, Nepal.
41
I realize that data
from the two studies are not directly comparable as they were from
different locations, but the point is that both parasite intensity and genetic
data are needed to fully elucidate the transmission process. Thus, in this
hypothetical comparison, although gender may account for differences in
worm burdens within a household (females have higher intensities
possibly due to peridomiciliary behaviors that increase exposure
32
), males
and females are still getting their worms from the same source of infec-
tion. Lastly, it should be noted that the patterns in Jiri may not extrapolate
to other locations as differences in human behavior, topography, and
external environmental conditions could alter transmission patterns. For
instance, a communal use of human feces for fertilizer may facilitate
parasite dispersal thereby creating a single source pool of parasites. Thus,
the assumption of a single infectious pool of parasites will need to be
tested for each population of interest and as evidenced by the study in
Jiri,
41
even on very local scales.
