Reader Comments
Post a new comment on this article
Post Your Discussion Comment
Please follow our guidelines for comments and review our competing interests policy. Comments that do not conform to our guidelines will be promptly removed and the user account disabled. The following must be avoided:
- Remarks that could be interpreted as allegations of misconduct
- Unsupported assertions or statements
- Inflammatory or insulting language
Thank You!
Thank you for taking the time to flag this posting; we review flagged postings on a regular basis.
closeReply to Hastings
Posted by plosmedicine on 31 Mar 2009 at 00:23 GMT
Author: Margaret Mackinnon
Position: Senior Scientist
Institution: KEMRI-Wellcome Trust Centre for Geographic Medicine Research, Coast
E-mail: mmackinnon@kilifi.kemri-wellcome.org
Additional Authors: Tabitha W. Mwangi, Robert W. Snow, Kevin Marsh, Thomas N. Williams
Submitted Date: March 09, 2008
Published Date: March 10, 2008
This comment was originally posted as a “Reader Response” on the publication date indicated above. All Reader Responses are now available as comments.
Hastings argues that because relatives share a common environment by living in the same house, or in the same household (i.e. a group of houses, just metres apart, occupied by an extended family), our estimates of the proportion of variance attributed to genetic effects (heritability) will be greatly over-estimated. We accounted for the shared common environment of households by fitting this effect in the analysis, but we did not fit an effect for house within household. The worry is that, as for household, house effects due to common environment may be mis-assigned to genetics, thereby inflating the heritability estimates.
Two aspects of our study mitigate against this. First, within each household, there was a mixture of full-sibs, half-sibs, cousins, uncles, aunts, etc. While full-sibs shared houses, half-sibs and other relatives did not. These non-sharing relatives, of which there were many in the data (typically three sets of second or third degree relatives in each household) can - if the appropriate analysis is done – disentangle the house and genetic influences. We made full use of extended genetic relationships within households in order to achieve this separation. In so doing we significantly advanced the field since, apart from this and our previous study in a Sri Lankan population [1], all previous studies on the heritability to malaria have made use of only a limited set of relationships, (e.g. twins or full-sibs), and/or their data structure has been such that house was totally confounded with sibship thus making genetics and house effects inseparable.
The second mitigating aspect was that houses within households were both spatially and socially very close to each other. Since the data were collected on the same children over a long period of time (5 years), people occupying the same household would, on average through time, share similar levels of exposure to mosquitoes, health-care resources and socio-economic status. It seems likely that differences between houses within households would be small, especially in comparison to between-household differences which dominated the total amount of variation. Nevertheless, we agree that the magnitude of between-house-within-household effects on disease susceptibility is an open question that requires further study.
We do not think it informative to use the standard errors of the estimates to test whether the heritability estimates are different from zero: all traits are to some degree heritable [2] and no such hypothesis was being tested here. (Had we wished to do so, however, we would have used the appropriate method, namely, a likelihood ratio test to compare the model’s fit against that under the null hypothesis). Certainly, the standard errors on our heritability estimates were large, as is characteristic for all heritability studies in natural populations. They are useful for reflecting the large amount of uncertainty in the estimates, but no more.
The number of malaria resistance genes being discovered grows steadily each year, and will now accelerate with the advent of large-scale genome-wide searches [3]. This suggests that there is a considerable amount of standing genetic variation in susceptibility to malaria, even after millennia of strong natural selection. Quite why this is the case is not known, but given the pathological nature of the immunological response to malaria, the high co-prevalence of other life-threatening infectious diseases that, like malaria, demand a balanced immune response [4], and the high degree of diversity in the parasite, it is perhaps not surprising that the host has no single, optimum, genetic way of dealing with malaria infection.
1. Mackinnon MJ et al. (2000) Proc Natl Acad Sci USA 97:12661.
2. Falconer DS, Mackay TFC (1996) Introduction to Quantitative Genetics, Longman.
3. Kwiatkowski DP (2005) Am J Hum Genet 77:192.
4. Graham AL (2002) Q Rev Biol 77:409.