AMOVA & MDS Calculation
To measure the apportionment of variance between pairs of populations, a method called AMOVA (Analysis of Molecular Variance) can be used (Excoffier L., Smouse PE., Quattro JM. (1992), ‘Analysis of molecular variance inferred from metric distances among DNA haplotypes: application to human mitochondrial DNA restriction data.’, Genetics 131(2), 479-91 [Pubmed]). In case of Y-chromosome profiles it considers the variance in the number of STR repeat units at DYS loci within and between populations. This method takes into account the molecular relationship of alleles, rather than just their frequency. The AMOVA method can be applied to measure the genetic distance between your population sample and reference samples from the YHRD.
By using the online AMOVA program it is possible to calculate FST or RST (apportionment of within/among population variance) between pairs of populations. To test for significance P values will be calculated as well (10,000 permutations).
Please perform online AMOVA as follows
- Use your population sample(s) file (Excel) to start the AMOVA. Be sure to provide haplotypes only for markers you are going to use in your AMOVA calculation. E.g. when using a Y23 file, all your AMVOVA is based on full Y23 haplotypes.
- Select Population Samples and/or whole National Databases from the YHRD. You can select up to 10,000 Haplotypes
- Check the calculation preferences and alter as needed
- Submit you calculation job and await the result
Contraints / Modifications applied to the set of haplotypes
- All haplotypes with empty or otherwise unspecified alleles (according to the markers given in the table header) will be removed.
- All haplotypes with alleles which do not comply with the markers nomenclature (e.g. “>24” or “OL”) will be removed.
- All haplotypes with intermediate alleles (e.g. “14.2”) will be removed.
- All haplotypes with null-alleles will be removed.
- All haplotypes with duplicated alleles (except DYS385 and DYF387S1) will be removed.
- All haplotypes with triplicated (or even more) alleles will be removed.
- DYS389I will be substracted from DYS389II.
The result will include
- Calculated pairwise FST or RST and p-values as CSV(Comma Separated Values) file.
- MDS(Multi Dimensional Scaling) as PDF(Portable Document Format) file. The MDS calculation is based on Kruskal’s non-metric MDS algorithm.
- Additionally, if “Relaxing” was selected: a relaxation file with all the clusters build prior calculation as text file.