I am working with a core collection or germplasm set which were sent for sequencing and now I have a couple of thousand SNPs markers. I intend to investigating the diversity. Explaining the diversity of the germplasm involves investigating the population structure. Apart from population structure and should I say Linkage disequilibrium within chromosomes, what other areas I also have to investigate and explain? and what are the procedures involve? Some Population Geneticist here could help?
hi dear Caleb Breria
you can also estimate PCoA, cluster analysis, allele frequency, and you used this data to association mapping.
This depends a lot of the nature of populations behind your germplasm set... for instance, providing the fact your collection is representative from a population, effective population size metrics (through LD for instance) can be interesting to compute...
Thank you Mostafa Haghpanah and Gregorie Leroy. Took note of both advice. On another note, in population structure, Structure Program (Prichard., et al 2009) was used to compute. While it gave me the best K value (sub-population), how do I figure out the best run within the best K subgroups so that the relevant Q-matirx file can be obtained for further computation. Take for example: if K was set @ K = 2 - 10 with 3 runs each. The result came out the best K = 2. How do I figure out which of the 3 runs within K = 2 is the best??
Again it depends on your data set. When confronted to this kind of issues, I used Clumpp software to identify the set of runs with the highest similarities G' and pick one of them, but for this you require much more than 3 runs per K.
Thank you Dr. Leroy, not that familiar with Clumpp software but am looking through the manual. regards
Dr. Leroy, had issues with the creation of .pop files for Clumpp. I then tried out the 'Pophelper' web app and it gave reasonable diagrams for each runs of Ks. Was wondering what is the (G' ) that you referred to above? Is it the ancestral matrix that you're referring to?
No I am referring to the similarity matrix, which is provided in the miscfile (see also Clumpp manual https://rosenberglab.stanford.edu/software/CLUMPP_Manual.pdf). It displays the similarities between run Q-Matrices. Based on such matrix you can see whether some of you runs provides similar patterns, and then chose the most relevant (frequent) ones...
- Badges
- Science topic
- Similar topics
- Psychology
- Mental Processes
- Learning