It's important to understand what Arlequin is doing for MANOVA. It will compare two or more populations to assess whether there is a difference in the allele frequency between the populations. It will also assess "interaction" (interpreting that in "gene flow" needs some thinking).
The basic theory is that if two or more populations cannot be distinguished under a formal parametric test there must be very strong gene flow supporting the lack of population structure between these populations.
Geographic foci Population designation is usually done between geographic foci, thus population A is usually geographic location (focus) A and population B is a geographically separated region (focus). Thus the fasta header needs to know the geographic region of the sample. Thats it.
Population info The issue is that information is required within the fasta header. Thus population delineation is therefore essential. This is confirmed in the PGDSpider manual below the line for Arlequin: which you don't need to do because thats what the conversion does.
Arlequin therefore needs to group each member of the population into a (usually geographically seperated) population to perform MANOVA. It will require this information specifying on the fasta header line. At the moment the data is saying there is a single population, it thinks all the sequences are from the same geographic location.
Procedure Thus it's likely simply a case of assigning two or more broad geographic foci for your samples. Then going through the fasta file and designating each sequence into one of the geographic focus under the keyword "population".
Just modifying @Steve's reading of the manual
> CcHr1 population:geoFocusA gene:X
ATAAAGATATTGGTACGTTGTATATATTATTGGGTGTGTGGTGTGGTATAGTTGGTACAGGATTATCATT
>CcMath1 population:geoFocusB gene:X
ATAAAGATATTGGTACGTTATATATATTATTGGGTGTGTGGTGTGGAATAGTTGGCACAGGGTTGTCATT
Then run PGDSpider
java -Xmx1024m -Xms512m -jar PGDSpider2-cli.jar -inputfile examples\example_Structure.fasta -inputformat FASTA -outputfile examples\output_Arlequin.arp -outputformat ARLEQUIN -spid examples\Structure_Arlequin.spid
You did attain the critical anomaly in the sequence format,
- each line of a sequence should have fewer than 80 characters
Your rows have 70 residues.
Genetic structure (only required for AMOVA):
The genetic structure specifies the hierarchical genetic structure of the samples. It is possible to define groups of populations.
o start of the subsection: [[Structure]]
o name for the genetic structure (string within ” ”): StructureName = “name”
o number of groups defined in the structure (int value): NbGroups = 5
o group definitions (list containing the names of the samples belonging to the group,
entered within braces “{ }”):
Id1 2 ACTCGGGTTCGCGCGC # the first pseudo-haplotype ACTCGGGCTCACGCGC # the second pseudo-haplotype
NbGroups=2
Group ={
population1
}
Group ={
population2
population3
}
Advantage The advantage of MANOVA is you can have LOADS of populations and it will assess which have high levels of gene flow between them. Statistically it is fantastic for loads of different reasons (quite complicated reasons). The disadvantage of output of MANOVA is there will likely be significant differences between many of the populations - possibly all of them. This is a result of the power of parametric statistics, but may not be particularly helpful for the analysis. If that is the outcome switching to Fst is the next part of the analysis (and a different question). Goodluck and you can always post back if you get stuck.
