As Vincent Ranwez says above, the sum of the branch lengths from the most recent common ancestor gives you the distance betwen the two tips. But the phylogenetic distance can be calculated in many different ways so that the roughly 0.26 distance you have between G and I in this tree (based on eyballing it from your scale bar) is probably not close to 26% DNA sequence difference (74% pairwise DNA sequence identity between the two). The exact percent identity between the two is call the "Hamming distance" or percent DNA sequence identity. The phylogenetic distance is corrected for many features of evolution, such as the fact that transitions happen much more frequently than transversions, and takes into account the issue of multiple hits per site.

To get the pairwise DNA distance you can use the multiple sequence alignment from which the tree was built, or you can use an alignment of just the two sequences. The result should be nearly identical unless the alignment was very problematic. Most phylogenetics and multiple sequence alignment packages provide a menu item for pairwise identity.

https://www.dnastar.com/megalign_help/index.html#!Documents/calculationofpercentidentity.htm

http://www.mbio.ncsu.edu/BioEdit/bioedit.html

https://www.biostars.org/p/70638/

Hi,

you can use the patristic distance, i.e. the sum of the lengths of the branches that link G and I to measure distance on the phylogenetic tree (see for instance Article PATRISTIC: A program for calculating patristic distances and...

If you need a similarity and not a distance, there are several solutions to turn a (patristic) distance into a (patristic) similarity (see for instance https://stats.stackexchange.com/questions/158279/how-i-can-convert-distance-euclidean-to-similarity-score)

As Vincent Ranwez says above, the sum of the branch lengths from the most recent common ancestor gives you the distance betwen the two tips. But the phylogenetic distance can be calculated in many different ways so that the roughly 0.26 distance you have between G and I in this tree (based on eyballing it from your scale bar) is probably not close to 26% DNA sequence difference (74% pairwise DNA sequence identity between the two). The exact percent identity between the two is call the "Hamming distance" or percent DNA sequence identity. The phylogenetic distance is corrected for many features of evolution, such as the fact that transitions happen much more frequently than transversions, and takes into account the issue of multiple hits per site.

To get the pairwise DNA distance you can use the multiple sequence alignment from which the tree was built, or you can use an alignment of just the two sequences. The result should be nearly identical unless the alignment was very problematic. Most phylogenetics and multiple sequence alignment packages provide a menu item for pairwise identity.

https://www.dnastar.com/megalign_help/index.html#!Documents/calculationofpercentidentity.htm

http://www.mbio.ncsu.edu/BioEdit/bioedit.html

https://www.biostars.org/p/70638/

I very much agree with the comment by Brian Foley that "To get the pairwise DNA distance you can use the multiple sequence alignment from which the tree was built, or you can use an alignment of just the two sequences. The result should be nearly identical unless the alignment was very problematic".

I will recommend the use of MEGA (i.e. Molecular Evolutionary Genetics Analysis) software which is free and provides a robust platform with tools for the computation of genetic distance matrices of aligned sequences. The results are presented in a number of easy-to-read formats, including Excel spreadsheets.

Hi Dr Deji

I think the thesis PDF file in the following link will more helpful in answering your question :

http://dspace.unive.it/bitstream/handle/10579/7695/834125-1188110.pdf?sequence=2

Yours