I have to cover around 10 sampling sites for answering hybridization between wolves and dogs and the genetic structure of wolves in India.
Well, my experinece indicates that if you have 10 samples from the site for genetic analysis you can get one picture, but rising it to 30 samples can significantly change the results. However, if you ask, for example, does hybridization take place or not and within the fisrt 3 samples you discover hybids, then 3 samples would be quite enough ;-)
The minimum for statistical purposes would be 30 samples for each site but it will depend on the polymorphism of the molecular markers that you will use.
Concerning this article, I remember the same approach (to detect the "most important traits" ) was very popular in the field of mtDNA research. And it is ok, if you can provide random sampling, but when concerns wolves and dogs and hybrids random sampling is hardly possible, because of socialilty and also because animals (and packs) are territorial. So, good sampling of various parts of study area could be more important than big number of samples
Hi Srinivas,
I agree with the previous comments above regarding sampling size, but especially with that of Nickolay Markov. As you have packs, your study design is nested and you have to modify your sampling design to this feature. From that point of view I would suggest to reduce the number of samples per location and increase the number of locations. Given the large terretories wolves could inhabit, you might have problems to come up with 30 samples at one location when there is just one pack of wolfes around and you probably have to sample each wolf twice or more often to get the n = 30, which makes of course no sense at all.
So, I would suggest to sample packs instead of locations. Try to sample ca. 5 individuals per pack (depending on how big packs are in your study region). Given the previous comments (30 samples for 10 locations) would result in 300 samples in total. Changing the design based on packs instead of locations and using 300 samples could allow you to sample 60 packs! Spread this number equally over your study area involving areas close to human settlements (likely more hybridization with feral dogs) and in the wild far apart from human settlements (if possible; hybridization less likely), then you would have a very nice and sound sampling design.
Also think about your marker system. If you work with microsatellites, make sure that you know in advance that you have private alleles, at least in some of the markers, to be able to determine possible hybrids. You should have a minimum of eight (better more than ten) loci that are working. If you haven't done so, make an initial screening on markers - that safes money and time.
Hope that helps.
Good luck,
Jan
Thank you all for the suggestions.
Hi Iris and Rene,
I will be using 25-30 maternal, paternal and bi-parental markers to address my questions. I am testing with more markers and finally will stick to 25-30 markers based on the success in amplification and polymorphism of markers.
Hi Jan, Thank you for writing in detail. But my question is what if there is no literature available on population estimates and pack sizes of wolves in India?
Hi Srinivas,
Pack size mostly depends on the type of prey the pack is specialized to or is available to hunt. In the high arctic pack sizes can easily exceed 15-20 members when hunting at big prey such as moose. In lower latitudes pack sizes normally decrease with the size of the prey and reach 5 members or so.
maybe you can use these studies as starting point:
http://onlinelibrary.wiley.com/doi/10.1111/j.1523-1739.1991.tb00354.x/abstract
http://onlinelibrary.wiley.com/doi/10.1111/j.1469-7998.2006.00103.x/abstract
cheers,
Jan
All the comments above are very valuable but I was also thinking that because you might get into trouble trying to find out what is going on, you might want to look at relatedness analyses package to see relationships among your packs. it might help out... and will definitively tell you if you have sample one individual twice ;-)
- Badges
- Science method
- Similar topics
- Biological Science
- Ecology