Deconstructing an infamous extinction crisis: Survival of Partula species on Moorea and Tahiti

Abstract

Eleven of eighteen Society Island Partula species endemic to the Windward Island subgroup (Moorea and Tahiti) have been extirpated by an ill‐advised biological control program. The conservation status of this critically endangered tree snail radiation is of considerable import, but is clouded by taxonomic uncertainty due to the extensive lack of congruence among species designations, diagnostic morphologies, and molecular markers. Using a combination of museum, captive, and remnant wild snails, we obtained the first high‐resolution nuclear genomic perspective of the evolutionary relationships and survival of fourteen Windward Island Partula species, totaling 93 specimens. We analyzed ~1,607–28,194 nuclear genomic loci collected with the double digest restriction‐site associated sequencing method. Results from phylogenomic trees, species estimation, and population assignment tests yielded monophyly of the Windward Island subgroup. Within this group, two well‐supported clades encompassing five species complexes were recovered. Clade 1 was restricted to Tahiti and contained two species complexes: “ P. affinis ” (three species) and “ P. otaheitana ” (five species). Clade 2 occurred on Moorea and on Tahiti and consisted of three species complexes: one Tahitian, “ P. clara / P. hyalina ”; the other two, “ P. taeniata ” (three species) and “ P. suturalis ” (six species), Moorean. Our genomic results largely corroborated previous mitochondrial DNA survival estimates for Moorea and Tahiti, with all five species complexes having members surviving in captivity and/or as remnant wild populations, although the details vary in each case. Continued, proactive conservation and management may yet ensure a phylogenetically representative survival of the fabled Partula species of Moorea and Tahiti.

Phylogenomic analyses confirm a novel invasive North AmericanCorbicula(Bivalvia: Cyrenidae) lineage

“…The DNA of the 36 Corbicula specimens genotyped in this study was extracted and their quantities assessed following the protocols of Haponski, Lee & Ó Foighil (2019). We targeted ~200 ng of DNA for library preparation, any individuals with DNA quantities less than this were re-extracted.…”

“…We targeted ~200 ng of DNA for library preparation, any individuals with DNA quantities less than this were re-extracted. ddRADseq libraries then were prepared and followed the protocols of Peterson et al (2012) and Haponski, Lee & Ó Foighil (2019). Prepared ddRADseq libraries were submitted to the University of Michigan’s DNA sequencing core () and run in two different lanes using 125 bp paired-end sequencing on Illumina NextSeq using the mid-range 300 cycle and a HiSeq 2500.…”

“…Raw sequences then were deposited on the Flux high computing cluster at the University of Michigan’s Center for Advanced Computing for further processing and analyses. The alignment-clustering algorithm in ipyrad v.0.7.17 (Eaton, 2014; Eaton & Overcast, 2019) was used to process and identify homologous ddRADseq loci using settings from Haponski, Lee & Ó Foighil (2019). Processed reads were clustered and aligned across individuals, filtered for paralogs, and finally concatenated into consensus loci at 85%, 90%, and 95% similarity de novo in ipyrad.…”

“…To determine phylogenomic relationships among the 36 Corbicula clams, we used methods detailed in Haponski, Lee & Ó Foighil (2019). We analyzed the nine concatenated ddRADseq alignment files of homologous loci using maximum likelihood (ML) in RAxML v8.2.8 (Stamatakis, 2014).…”

“…We again specified the two C. japonica individuals as the outgroup. Bayesian analyses followed Haponski, Lee & Ó Foighil (2019) and included the GTR model with invariable sites and a gamma distribution and used a Metropolis-coupled Markov chain Monte Carlo (MC 3 ) approach and ran for 4,000,000 generations, with sampling every 100. Two analyses were performed each with four separate chains run simultaneously.…”