The paper in Nature Ecology & Evolution is here: https://go.nature.com/2GswrVL
The idea behind dating with transfers is simple: a transfer can only happen between two contemporary organisms. In a phylogenetic context this means that the ancestors of the gene donor (blue) must be older than the descendants of the recipient (red). This is a simple and elegant idea first suggested by J.P. Gogarten but it has never been shown to work in real life.
When I started my PhD at the LBBE in Lyon I got the chance to work with the scientists who develop the methods necessary to put these ideas into practice. Ideally one would compare directly dates inferred from the fossil record and the time information from transfers. Unfortunately, it appears to be the case that where there are fossils there are no transfers and vice-versa. The idea for the main result of the paper came when we figure out that we could compare Molecular Clocks (the standard way of dating phylogenies) with the relative ages implied by transfers.
This took some time to put into practice but when I finally managed to compare the relative orders inferred with the transfer events with those inferred by molecular clock we were surprised to discover that there was a high degree of similarity between them.
My supervisor's first reaction (and second, and third...) to this apparent correlation was to think of various ways that it might be wrong. We spent more than a year and a half trying to find how it would be possible to obtain a similar result if the signal was spurious. After failing in this enterprise of debunking our own results, we reached the conclusion that the signal was genuine and that transfer events indeed carry a strong dating signal. I learnt in the process that in science you have to spend about 80% of the time just making sure you are not fooling yourself.
Our results provide a convincing demonstration that transfers seem to have a dating signal that agrees with clocks. That said, the methods are in many ways in their infancy, and we need new developments, particularly in combining the signals from transfers and molecular clocks. This line of research is a promising path that will lead us hopefully to a better understanding of the evolutionary events that took place in a distant time.
Image credit: Alba A. Davín
Please sign in or register for FREE
If you are a registered user on Research Communities by Springer Nature, please sign in
Shareable author link to full text here: http://rdcu.be/KrjA