Is it normal in Network 4.6 that the most frequent haplotype and other 2 disappear?.

Hi Alexandra

Why don't you use simply TCS for dealing with your chlorplast haplotypes (defined as the combination of the alleles obtained at each locus). Numerous examples in the literature were using it for the construction of haplotype network with cpSSR data (see attached some study cases)!!!

Nicolas Thiercelin

Just to be sure, do you have missing data (as question marks) in your dataset ?

Concerning TCS, there is a tool for it in the excellent Fabox: http://users-birc.au.dk/biopv/php/fabox/fasta2tcs.php

Alexandra Ricardo

I don't have missing data. What I have is the size of the cpSSR fragment instead of the complete nucleotide sequence. I'am going to try to use TCS and compare with the results of Network 4.6.

Thank you for the help.

Cheers

Nicolas Thiercelin

I went back to the papers building haplotype networks with cpSSR, and I noticed that at least two used a matrix of pairwise differences as dataset for the haplotype network. This method seems relevant as it counts the number of differences between your haplotypes, especially as your haplotypes are (if I'm right) not mutations but number of repeats.

The papers I'm referring are these ones:

http://onlinelibrary.wiley.com/doi/10.1046/j.1365-294X.2003.01890.x/full

http://onlinelibrary.wiley.com/doi/10.1111/j.1365-294X.2012.05619.x/full

You can also check if the method described in this study is relevant in regards with your dataset: http://onlinelibrary.wiley.com/doi/10.1111/j.1365-294X.2006.03064.x/full

Adriana López-Villalobos

Hi Alexandra… I know this post was from a while ago and you are probably not longer interested or have solved the issue. Anyway, I just came across with this and I was actually searching for something specifically related to an input file with cpSSR using this software.

I have also used the software with the same kind of data you have to infer an haplotype network.There is a default option that says that haplotypes with a frequency less than 1 (i believe) will be removed from the network. You can turn that option off and tell the software to include all the haplotypes even if they have a low frequency. That could be one thing. The other issue (as mentioned above) might be that the input data file is not exactly as it should be. This time I am actually having a hard time understanding how to make the input file since I only have fragment size for each individual at each locus, rather than the number of repeats (by sequencing). As you might already know the software manual is neither very helpful nor self explanatory.

I would be curious to know how did you deal with the input file. I am using the example data file of YSTR they provide. But I don't know whether one should use # of repeats as direct fragment size (230, 232, 233 bp which actually the program doesn't seem to like) or some sort of number of differences among alleles for each locus, considering I do not have the exact number of bases (by direct sequencing) by which each allele varies in each locus ( using universal cpSSR, not species-specific).

Anyway... If somebody here has an idea, I will greatly appreciate your input.

Thanks!

Problem with my Haplotype Network using Network 4.6

What degree of missing data in a particular case/participant results in Expectation-Maximisation no longer working?

Is There Any Feasible Method To Test Fluorescent Compounds Other Than Fluorescence Spectrometers ??

What is the difference between genotype and haplotype data in the ped format?

Mixed ANOVA - help?

How to Interpret Partial Eta Squared Numbers For a Repeated Measures Anova?

Interpretation of Goodman - Kruskal Gamma squared?

Interpretation of a fluorescence Emission spectrum ?

In CDA analysis we have three stages, viz., description , interpretation, and reproduction. Why have all CDA studies neglected the third stage?

How can I interpret a ratio variable?

Does the phylogenetic tree making tool IQTREE use evolutionary placement algorithm?