they interpret this as the parasite making the mosquito bite people, but seems like there's a problem with reverse causation: mosquitoes that bite people are more likely to get malaria
Show this thread
this preprint claiming to show that malaria manipulates mosquitoes into biting people sounds cool, but i don't understand the basics of their experiment. they collected mosquitoes & found that ones that had bitten humans were more likely to have malaria sporozoiteshttps://twitter.com/mbalter/status/922108794359791616 …
-
-
-
they do address this! key point is that mosquitoes with (infectious) sporozoites are more likely to have bitten human recently than those with (earlier, non-infectious) oocystspic.twitter.com/4PCWYgyE9e
Show this thread -
but i don't see how this solves the problem. in the very next line, they point out that P. falciparum takes ~1 mosquito lifetime to develop into sporozoitespic.twitter.com/q5molsByKT
Show this thread -
before that, it spends "several days" in the oocyst stage https://en.wikipedia.org/wiki/Plasmodium_falciparum#In_mosquitoes … so let's say you catch a 10-day-old mosquito. to have sporozoites, it must have bitten an infected person on day 1. to have oocysts, it could have bitten them on days 2-5
Show this thread -
the mosquitoes that bit humans more frequently would be overrepresented in the first category compared to the second, i.e., you'd see an association b/w biting humans and having sporozoites over oocysts even with no manipulation
Show this thread -
how this actually would work out depends on the exact numbers for malaria & mosquito life cycle & resting period b/w bites, which i don't know. could even be that it ends up working in the opposite direction of what i've sketched here! but i think it should be checked
Show this thread -
(particularly if one wants to use the estimated effect size for epidemiological modeling, as they do in the preprint)
Show this thread -
something struck me in their data that i don't think they comment on: they caught 184 mosquitoes with sporozoites & only 118 with oocysts (& 25 with both). if my speculation that mosquitoes spend more time infected with oocysts than with sporozoites is true…
Show this thread -
… i would have expected the opposite pattern. obviously the numbers are not huge & there are lots of possible explanations, but it could potentially be a sign of manipulation by the parasite
Show this thread -
but i guess they're really looking for signs that malaria is manipulating the choice of biting targets (human vs non) and there's no reason to think that it would be that kind of discriminating manipulation (i.e., could be just going towards CO2)
Show this thread -
this is a really good point https://twitter.com/mathiesoniain/status/1038162618324844545 … i'm leaning more towards the authors' interpretation now. thanks
@mathiesoniain!Show this thread -
on the third hand, their data *do* show mosquitoes with oocysts have an increased prob of having bitten a human than uninfected ones, it's just not significant. but not clear that we would expect it to be under a simple null model given the sample size. might need to write it out
Show this thread
End of conversation
New conversation -
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.