Kristian G. AndersenGeverifieerd account

“Other than fast computation and very attractive graphics, MJNs harbour no virtue for phylogenetic inference. MJNs are distance‐based, unrooted branching diagrams with cycles that say nothing about the evolutionary history due to the absence of direction. “...

...”MJ was introduced in 1999 and, in contrast to most scientific ideas, its application has spread rapidly through copying the methods of others, and, unfortunately, with little further scrutiny. We hope that the theoretical arguments presented here can reverse this trend.”

Let me play the devil's advocate: frequent sequencing (not the case here) can track mutations arising 1-by-1 & graph cycles can indicate homoplasies (not expected here) with the temporal dimension easy to add later, so I'd say MJ has potential as data-vis but there's no effort.

we have used these networks in the past, but with v closely related seqs following a sweep w/ star shaped tree - we also presented proper trees, but when you have lots of samples w/ ID seqs then others seperated by v short branches (in our case 1-3 mutations) then ...1/2

... perhaps these can be easier to interpret...? But I may be wrong. In which case I blame my internal PhD examiner for not pointing this out in my thesis. Eg above is bit bonkers though, clearly not good use!

There are reasonable use cases - to try and reconcile incompatible reconstructions (due to selection or recombination) but using it as a reconstruction of a transmission tree is probably a very bad idea. I feel a simulation experiment coming on...

There are cases where I think these networks make sense - densely sampled, resolved, closely linked virus transmission chains - aka simple scenarios. That's not what is going on here - and of course, trees will *always* be more informative (but yes, require skills to interpret).