WiSPA: A new approach for dealing with widespread parasitism

Abstract

Traditionally, studies of coevolving systems have considered cases where a parasite may inhabit only a single host. The case where a parasite may infect many hosts, widespread parasitism, has until recently gained little traction. This is due in part to the computational complexity involved in reconstructing the coevolutionary histories where parasites may infect only a single host, which is NP-Hard. Allowing parasites to inhabit more than one host has been seen to only further compound this computationally intractable problem. Recently however, well-established algorithms for estimating the problem instance where a parasite may infect only a single host have been extended to handle widespread parasites. Although this has offered significant progress, it has been noted that these algorithms poorly handle parasites that inhabit phylogenetically distant hosts.

Strong phylogenetic congruence betweenTulasnellafungi and their associated Drakaeinae orchids

“…We used the default cost model (zero for cospeciation, one for duplication, two for duplication and host‐switch, one for symbiont losses, and one for failure to diverge) (Conow et al, 2010 ). A spread event is common among pathogens that can infect multiple phylogenetically distantly related hosts (Drinkwater et al, 2018 ). However, spread is not covered by current event‐based methods due to its computational complexity (Charleston & Libeskind‐Hadas, 2014 ; Drinkwater et al, 2018 ).…”

“…Our results also showed that there were nine failure to diverge events, where the orchid has failed to speciate in parallel with divergence in the fungi. However, an alternative explanation, which is not captured in current event-based analyses due to computational difficulties (Charleston & Libeskind-Hadas, 2014;Drinkwater et al, 2018), is that the orchids have experienced "spread" events (Drinkwater et al, 2018;Fraija-Fernández et al, 2016), whereby orchid species have retained their original fungal species but then also Host-switching occurs when a duplication event of the orchid is followed by one of the two sister orchid species switching to a fungus that belongs to a different clade in the fungus phylogeny (Charleston & Perkins, 2006;Conow et al, 2010). Host-switching is not prevalent for Drakaea and Chiloglottis (no events identified), but appears to have occurred more frequently for the fungal associations of Arthrochilus (two events).…”

“…In the case of orchids, the dependency of the orchid on the fungus, but not vice versa , means that in these analyses the orchid takes the position of a symbiont, and the fungus the host. In host–symbiont cophylogenetic studies, six macroevolutionary scenarios may be observed; (i) cospeciation/codivergence, where host and symbiont speciate simultaneously, ii) duplication, where the symbiont speciates and the resulting sister species remain using the same host, (iii) host‐switch, where speciation is accompanied by one of the descendants transitioning to a different host, (iv) loss, where the symbiont fails to track one of the new lineages of the host due to extinction or the host colonizes a new habitat without the symbiont, (v) failure to diverge, where the symbiont associates with both hosts following host speciation (Charleston, 1998 ; Charleston & Perkins, 2006 ; Conow et al, 2010 ), and (vi) spread, which consists of a symbiont colonizing an additional host without the symbiont speciating (Charleston & Libeskind‐Hadas, 2014 ; Drinkwater et al, 2018 ). A pattern of strong phylogenetic congruence can arise either from frequent cospeciation or if host‐switches typically involve transitions to closely related host species (Hayward et al, 2021 ).…”

Cophylogeny Reconstruction Allowing for Multiple Associations Through Approximate Bayesian Computation

“…To the best of our knowledge, the parsimony-based methods that address multiple associations are the following: CoRe-Pa (Merkle et al, 2010), Jane 4 (Conow et al, 2010) and WiSPA (unpublished, see Drinkwater et al (2016)). The tool CoRe-Pa (Merkle et al, 2010) deals only with the case of cryptic species and solves the multiple associations locally in a parsimonious way.…”

“…The tool CoRe-Pa (Merkle et al, 2010) deals only with the case of cryptic species and solves the multiple associations locally in a parsimonious way. Jane 4 (Conow et al, 2010) and WiSPA (Drinkwater et al, 2016) consider multiple associations as resulting only from recent host switches (case (iii) above, see Brooks et al (2004)).…”