Yes, you can estimate times of divergence from a common ancestor by rates and distances.  However, it is best to use exactly the same gene regions and methods on both your data set of unknown dates and the data set with nodes known from the fossil record.  The distances in most phylogenetic trees are not simple % distance, they are usually "weighted" distances which take into account that transitions happen far more often than transversions, and multiple hits per site can accumulate at larger distances.

For example, HIV-1 is observed to evolve at roughly 0.5% per year, to the distance between two viruses that shared a common ancestor one year ago is 1%, one change per 100 bases, because each virus changed 0.5% from the common ancestor.  After 100 years from a common ancestor the viruses are not 100% distant, any two random strings of AGTCGATGC will share 25% identity by chance.  After 100 years from a common ancestor HIV-1 sequences will be closer to 75% identical to each other because of selection pressures etc.  Phylogenetic reconstruction attempts to account for this non-linear change in distance over time.

Different genes evolve at different rates and different sites within a single gene evolve at different rates, so it is best to compare exactly the same data for the species with known and unknown dates of divergence.

Yes, you can estimate times of divergence from a common ancestor by rates and distances.  However, it is best to use exactly the same gene regions and methods on both your data set of unknown dates and the data set with nodes known from the fossil record.  The distances in most phylogenetic trees are not simple % distance, they are usually "weighted" distances which take into account that transitions happen far more often than transversions, and multiple hits per site can accumulate at larger distances.

For example, HIV-1 is observed to evolve at roughly 0.5% per year, to the distance between two viruses that shared a common ancestor one year ago is 1%, one change per 100 bases, because each virus changed 0.5% from the common ancestor.  After 100 years from a common ancestor the viruses are not 100% distant, any two random strings of AGTCGATGC will share 25% identity by chance.  After 100 years from a common ancestor HIV-1 sequences will be closer to 75% identical to each other because of selection pressures etc.  Phylogenetic reconstruction attempts to account for this non-linear change in distance over time.

Different genes evolve at different rates and different sites within a single gene evolve at different rates, so it is best to compare exactly the same data for the species with known and unknown dates of divergence.

Hello Yusdel,

you can do this in *BEAST. I found this post in the BEAST2 website very useful to set things correctly in BEAUTi.

http://beast2.org/2015/06/23/help-beast-acts-weird-or-how-to-set-up-rates/

Hope this can help you,

Best,

BEAST produces so called ultrametric tree with branch lengths reflecting the genetic distance. You can convert this with known mutation rate to absolute time. (But see the warnings above). The method is quite simple, in your case take the relative node age e.g. 0.35 than the absolute age will be 14 mya (0.35 x 1000000 / 0.025). However this is a very rough estimate use it with care!

Csaba, Thank you so much for your answer!!

Dear Csaba

how did you detect the 0.35 relative node age and the equation parameter (0.35 x 1000000 / 0.025)

I am working in mtDNA of toad and my publication is already missing only this point for publication.

how can I detect the mutation rate from BEAST.

Thanks

You cannot infer the substitution rate (not mutation rate) from a BEAST analysis just if you have one (desirably more) external calibration point. If you have an already published substitution rate for a sequence in your taxonomic group than you can convert the relative node age provided by BEAST to an estimated absolute age. Also there is a possibility to run the BEAST analysis using with the known substitution rate.

Dear Csaba

I am interested in your statement above "If you have an already published substitution rate for a sequence in your taxonomic group than you can convert the relative node age to an estimated absolute age".

Could you articulate a bit more on how to do this? I see you provided an example above (0.35 x 1000000 / 0.025). What do these numbers correspond to in the equation?

Thank you for help