Entering edit mode
5.5 years ago
nitishd.dave
▴
20
Can anyone suggest me what to do if a single individual is completely going out in the dendrogram (generated from RAPD markers ) from the rest of the population even if the species is same and a sample is also collected from the same area? I mean if that sample should be included in the dataset or not and if not is there any literature supporting this argument.
You mean you have one extremely long branch when you should expect it to group tightly with the others?
Is this generated from a sequence alignment? Have you checked the sequence alignment looks trustworthy?
One anomalous branch would usually suggest to me that sequence is incorrect.
thank you for the reply, Yes indeed a long branch is obtained however, it is not a sequence data the cladogram is generated using a binary matrix from a PCR gel image (RAPD MARKER). should I remove this one branch from the dataset and if there is any literature supporting this theory.
Have you checked the gel image? I’m not familiar with the RAPD technique, but one way or another it sounds like an experimental failure of some kind.
Are the branch lengths derived from band intensity in that case?
I imagine it would be better to remove it than include incorrect data, but I suppose it depends what you wish to do with the tree. There is something to be said for leaving the data in for transparency and just adding a caveat that that branch may be an experimental outlier.