Hello Yousef,

unfortunately you cannot do that because you basically don't know the real evolution rate of your sequences, and evolution rate are known to vary through time. In special cases you can infer it from the estimates of other phylogenetic studies on related groups, as they did here:

Article The towering orogeny of New Guinea as a trigger for arthropo...

Another solution would be, for instance with nematodes (your focus if I'm right), to say that if you have a monophyletic group of nematodes using the human as exclusive host, then you can make the assumption that this group is younger or as old as the human, and this way you have a calibration point in your phylogeny.

Cheers,

Théo

It depends a LOT on what type of organism you are working on, and what timescale you are interested in. Clams and mussels today look a lot like the fossils from 400 million years ago, so determining speciation events even with the fossils can be difficult. Most mammals have changed morphology quite a bit in the past 1 million years or less. Mitochondrial genomes evolve roughly 10X faster than nuclear genes and are inherited usually only from eggs not sperm, while most nuclear genes recombine over time and are found in both sperm and eggs.

Viruses and bacteria never leave fossils that are identifiable at the species level, but there are often other dated events in their histories from which we can infer some time:genetic distance relationship.

Even if you cannot exactly specify the time:distance rate, you can get a general ballpark estimate from most data sets, if you mine GenBank for data set related to yours where there is some known dated evidence.

Many thanks for your attention and answers. Iam working on the phylogenetic study of plant parasitic nematodes and unfortunately this group of animals have any fossil records.

Do not use "secondary calibration dates" = the results of other divergence-time analyses. If you do that, the uncertainties of your analysis will multiply themselves with the uncertainties of the earlier analysis, and soon you'll end up with error margins as long as the Phanerozoic.

If the parasites are host-specific enough, starting from the timetree of the hosts is probably your best bet, but it's still dangerous, unfortunately.