3 ms·
Such gaps in phylogenetic trees actually happen all the time.
by epgui 5y ago
Such gaps in phylogenetic trees actually happen all the time.
- TeeMassive 5y agoOther viruses are not scanned and studied for other variants like Covid-19 is and Omicron is the only outlier by a large margin.
- zarzavat 5y agoSure and if it had arisen in Europe then those questions would have to be asked. But it's a safe assumption that very little analysis is happening in southern central Africa. It doesn't seem that outrageous that a parallel line has been percolating there for quite some time and only recently gained a mutation or two that allowed it to shoot to stratospheric levels.
- epgui 5y agoEven in Europe I wouldn't find it all that difficult to believe. We see the same sort of pattern, qualitatively speaking, even in well controlled in vitro experiments.
- rleigh 5y agoGaps can appear if there are (1) missing data or (2) a big change such as a transposition or elision event. Neither are the case for our coronavirus data. For (1) we have a good random sampling of the population as a whole so it is exceedingly unlikely we would not see intermediate changes and for (2) the overall structure has not changed dramatically--we are looking at the accumulation of point mutations over the course of 18 months. So while gaps can and do occur, we would not expect this to occur naturally in the situation we are observing today with coronavirus.
- epgui 5y agoGaps also appear because evolution is not a smooth process, and once in a while a large number of mutations are selected for. We see the same type of pattern even in vitro. Moreover, mutations are not always well described by gaussian processes: the rate of mutation can be seen as having a fat tailed distribution.
- TeeMassive 5y agoThat's not the way it works. Mutations has to be incremental because most mutations are not viable or doesn't provide any reproductive gain. Having multiple functional mutations all at once happening in a very short amount of time and being naturally selected in big enough number of infections is very very improbable. > mutations are not always well described by gaussian processes: the rate of mutation can be seen as having a fat tailed distribution I'm sorry if I have to be blunt, but you have no idea of what you are talking about.
- epgui 5y agoI’m a biochemist.