Help in understanding phylogenetic trees of Gastropod

Nature Talk
1

Observation - https://www.inaturalist.org/observations/209642403

I am just curious and no expertise in phylogenetics.

context Purpura panama vs persica vs rudolphi: rudolphi is synonymised with one species in some and another in some resources historically. Indian westcoast rudolphi variant is recently COI sequenced and I picked the COI sequences of other two from BOLD.

I used BLAST first and then used NGPhylogeny as I read it offers better statistical support than BLAST.

BLAST results: https://web.archive.org/web/20251210140351/https://blast.ncbi.nlm.nih.gov/Blast.cgi?CMD=Get&RID=KJWVKKA7114

image.png

NGPhylogeny results: https://ngphylogeny.fr/workspace/history/acf1d4b25e6e2c47

image.png

so with these results, my understanding is that Indian rudolphi could be aligned with panama better?
or rather new species? or whether the current tree is messy to conclude anything concretely and needs more samples? or there is need for change in my method above? can anyone help for this case?

On sidenote, the last phylogenetic study on subfamily did separate panama and persica (although i dont know what exact COI sequences they used for this two) and mentioned Purpura bufo is uncertain of genus placement with lack of more genus samples of this - https://www.sciencedirect.com/science/article/abs/pii/S1055790312003715

2

I have limited experience with building trees, but I noticed that you didn’t include an outgroup (a species outside the group of interest). This is necessary for properly rooting the tree.

Also, using single genes is not always reliable for gastropod phylogenies. In particular, mtDNA like COI has been shown to be highly variable even within species:

https://doi.org/10.1186/s12864-024-10505-w

1 Like
3

oh yes thanks, i was mistakenly assuming the differences in ingroup can provide tree with some differences like Indian rudolphi not in persica clade, but yes now i see outgroup is essential i guess as with rooting nodes that may not fit to any clades confidently within genus.

ahh yes, i think multi-gene phylogenetics are modern standard for confident delineations and cryptics. The last paper I linked above also used four genes for their trees instead of only COI in this subfamily.

We only have COI seq for indian one as of now, and I read this line in the paper:

Genetic clusters were defined as monophyletic groups where COI divergence was less than 6%, following previous work on other rapanine species - Claremont, M., Williams, S.T., Barraclough, T.G., Reid, D.G., 2011b. The geographic scale of speciation in a marine snail with high dispersal potential. J. Biogeogr. 38, 1016–1032.

similar results have been shown in other gastropods - Reid, D.G., 2007. The genus Echinolittorina Habe, 1956 (Gastropoda: Littorinidae) in the Indo-West Pacific Ocean. Zootaxa 1420, 1–161. Reid, D.G., 2009. The genus Echinolittorina Habe, 1956 (Gastropoda: Littorinidae) in the western Atlantic Ocean. Zootaxa 2184, 1–103.

If this is the case, and how COI of Indian one has 8%ish difference in BLAST alignment (first image result) from other two, is it safe to assume that this is neither of those species clades then? (assuming the single Indian COI seq isnt too erratic)

4

You may be able to assume that, but that also depends on things like query cover. In Cepaea there are differences of >10% within species, which is unusually high. I don’t know if this could also happen in marine species.

Either way, using a genomic approach like RADseq or UCEs is better. But it’s best to have a whole genome sequenced first, which might not be the case for this group. You should consider submitting a nomination for mollusk of the year to have its genome sequenced! I believe they may prioritize nominations from people/institutions/countries that don’t have as many resources for this.

https://docs.google.com/forms/d/e/1FAIpQLSdSI5SQV4kYwOGArJyAmcH5SXdnUDUJ8sW0NVqi90aEFLUlHg/viewform

1 Like
5

super cool. thanks.