Anchored Hybrid Enrichment and the prodigious clavaeformis confusion

Editor’s Note – This is the second installment of a projected three-part series reviewing the 2022 paper by Whelan and colleagues [1] on “Prodigious polyphyly in the Pleuroceridae.”  We recommend that you back up and read last month’s post [23Sept25] before proceeding onward, if that material isn’t fresh in your mind.  You will find a review of our August post [20Aug25] helpful, as well.  And in fact, refreshing your memory on our posts of September – November of 2016 [2] will also fill in a lot of the relevant background.  But that’s probably too much to ask.

Last month we summarized, to some modest level of detail, the fancy new technique called “Anchored Hybrid Enrichment” (AHE), recently wielded by a team of our colleagues headed up by Dr. Nathan V. Whelan of Auburn University to construct a phylogenomic tree of the North American Pleuroceridae.  AHE analysis begins with the development of “anchors,” DNA probes designed to target and enrich highly conserved regions within the genomes of the big set of organisms under study.  The slang term for these highly-conserved anchor regions is “bait.”

So, this month, let’s talk about bait.  Nathan and his team selected five probe species to “maximize phylogenetic diversity” across the Pleuroceridae, extracted mRNA from that small subset, and ultimately developed from the transcriptomes thereof 742 baits, requiring that they be present in at least four of the five probe species and larger than 120 bp in length.

Upon further review, I would not fuss about the identifications of three of those five probe species: Leptoxis “ampla” [3], “Elimia” crenatella [4], and Lithasia geniculata [5].  I could not find images of those three shells in the supplementary materials provided (USNM 1638590, 1638624, and 1638596, respectively) but I could find images of other nominally conspecific individuals bearing the same locality data, and they looked OK.

Pleurocera clavaeformis in Pistol Creek [6]

Let me set aside Nathan’s fourth probe species, “Pleurocera prasinata,” and come back to that situation later in this essay.  With regard to the fifth individual snail that the Whelan team selected for the generation of their AHE bait set, which they identified as "Elimia clavaeformis," what a mess.

Back in August [20Aug25] we reviewed the paper I published in 2011 [6] vividly demonstrating cryptic phenotypic plasticity in populations of that species, best identified today as Pleurocera clavaeformis, as they range across East Tennessee.  The shell morphology demonstrated by many upstream populations, historically identified as “Goniobasis (or Elimia) acutocarinata” is gracile with a striking peripheral carination.  The shell morphology of populations inhabiting mid-sized rivers and streams is what everybody has always associated with typical Goniobasis (or “Elimia”) clavaeformis.  And the shell morphology of big river populations, historically identified as “Pleurocera unciale,” is smoothly-rounded and robust.

Now look at Nathan Whelan’s Figure 5, as reproduced below.  Even given a healthy appreciation of the dramatic phenotypic variation in shell morphology that can be demonstrated by the pleurocerid populations of North America, do the four shells depicted across the top row of that figure look like the same species as the four shells depicted on the bottom row?  Who among my vast and far-flung readership cannot immediately distinguish two subsets of shells in Nathan’s figure: an ABCD group and an EFGH group?  Contact me immediately!  Your FWGNA credentials are in dire jeopardy.

The four shells in the EFGH group on the bottom row of Nathan’s figure below all belonged to individual Pleurocera clavaeformis acutocarinata, fine, as we just spent the entirety of our August post laboring over, no problem there.  But the pair of smooth, squat-looking shells in the upper right, labeled C and D, belonged to two Pleurocera simplex of very typical morphology.  And the pair of smooth, slender-looking shells in the upper left, labeled A and B, belonged to its sibling species, Pleurocera gabbiana.

“Figure 5. Shell morphology of Elimia clavaeformis s.l.” [1]

My readership can be forgiven for confusing simplex and gabbiana.  They co-occur in mixed populations throughout much of East Tennessee, and I myself was only able to distinguish them after rather tedious allozyme and morphometric studies, which I published in the FMCS newsletter Ellipsaria in 2016 [7], and relayed in three posts to this blog later that same year [2].

But alas, I bear bad news. A team of five professional malacologists, headed by our own colleague Nathan Whelan, identified all eight of the shells shown in Figure 5 above as belonging to “Elimia clavaeformis s.l.”

It is difficult for me to understand how the Whelan team missed my 2016 simplex/gabbiana series in Ellipsaria [8], near impossible for me to understand how they missed my 2011 paper on clavaeformis published in Malacologia, and inconceivable to me that they could confuse clavaeformis with simplex.  Yet that is exactly what they did.

Both shell A and shell F [9] were collected sympatrically at Turkey Creek, about 20 km W of Knoxville.  Both C and G were also collected sympatrically from Pistol Creek at Maryville, 25 km S of Knoxville, one of the four streams I myself sampled for my 2011 clavaeformis study, and also where I conducted my simplex/gabbiana study in 2016.  The reproductive isolation between the clavaeformis and simplex populations inhabiting those creeks is writ vivid on the shells of the animals that bear them, entirely distinctive, 100% sortable at streambank.  For five of our colleagues to confuse P. clavaeformis with P. simplex/gabbiana four separate times, cavalierly dismissive or abysmally ignorant of previously published research, and advertise their abject confusion in a full-page montage in an international scientific journal, is the most inconceivable concatenation of error upon error since Vizzini drank the goblet.

Now here is some good news.  Nathan’s supplementary material included a photo of the shell on the back of the individual snail from which he developed his AHE bait set, and it turns out to have indeed belonged to a Pleurocera clavaeformis acutocarinata.  So, his initial identification was correct.

And here is a second item of even better news.  Even if that snail had been misidentified, it would not have mattered.  To generate a good AHE bait set, all that really mattered to the Whelan team was that the five probe species they selected cover the range of genetic divergence across the 194 snails they intended to troll with them.  Which four of the five probe species certainly did.  Which almost brings us back to probe species #4 (of 5), which Nathan identified as “Pleurocera prasinata.”  But let’s put that issue off again.  And continue to move forward.

Seven of the eight snails depicted in Figure 5 above were included among the 194 test individuals in Nathan’s overall AHE analysis, and those seven did appear hanging among the branches in the big set of huge phylogenomic trees (generated under various assumptions) that he made available in his supplementary material.  Nathan only selected a couple of the Figure 5 shells to depict on the pretty tree he ultimately published as Figure 4 in the journal article, reproduced below.  But the bottom line is this.

Figure 4 of Whelan and colleagues [1]

Snails A and B clustered in the group I have labeled Pleurocera gabbiana and snails C and D clustered in the group I have labeled Pleurocera simplex.  Snail G (Pistol Creek) clustered in another big group a bit further around clockwise, with taxa Nathan labeled prasinata, acuta, pyrenella, attenuata, and so forth.  And snails E (Town Creek, not depicted) and F (Turkey Creek, depicted) clustered in a group much further around on the other side of the tree, between a snail Nathan labeled arachnoidea and a snail he labeled edgariana.

Knowing what we know about the biology of these remarkable organisms, the placement of the Pistol Creek individual is correct.  Look back up at the top of this essay, to that detail from Figure 2 of my 2011 paper.  As I mentioned above, the Pistol Creek P. clavaeformis population was one of the four I selected for allozyme analysis in that research.  My sample CA2 was taken from the exact spot that Nathan collected his individual Pistol Creek snail, an upstream reach running through a lovely little park in downtown Maryville.  I also sampled the Pistol Creek P. clavaeformis population 8 km further downstream, where it demonstrated the typical shell morphology depicted as C2.  And the shell labeled P2 was collected another 6 km downstream and out into the Little River.  That robust P2 shell phenotype, historically identified as “Pleurocera unciale” throughout East Tennessee, is indistinguishable from the shells Nathan has identified as “prasinata” and “aff. acuta” in his Figure 4 above.

And now at last, we see the problem with Nathan’s selection of both “Elimia clavaeformis” and “Pleurocera prasinata” as probe species from which to develop his AHE baits.  Far from “maximizing phylogenetic diversity,” those two nominal species appear so genetically similar as to be conspecific [10].

So now let’s turn our attention to that pair of P. clavaeformis acutocarinata (only one of which was figured) collected from Turkey Creek and Town Creek clustered way off in the upper left quadrant of Nathan’s tree with arachnoidea and edgariana.  Science is the construction of testable hypotheses about the natural world.  Let’s do some science.

In a series of essays posted on this blog late last year I reviewed what we know about interspecific hybridization in the North American Pleuroceridae, documented widespread hybridization between Pleurocera laqueata and P. troostiana in Middle Tennessee, North Alabama, and Kentucky, and suggested that reticulate evolution may have played a significant role in the diversification of the family across eastern North America generally [11].

The nomen “Elimia arachnoidea” depicted immediately above the Town/Turkey acutocarinata in Nathan’s Figure 4 is a junior synonym of Pleurocera troostiana, and the nomen “Elimia edgariana” depicted immediately below is a troostiana/laqueata hybrid [12].  The “Elimia sp. Red River” below edgariana is unidentifiable, but next below Red River is “Elimia curreyana,” a junior synonym of Pleurocera laqueata.  My hypothesis is that both of the individual snails sampled from Town Creek and Turkey Creek are clavaeformis/troostiana hybrids, a possibility I mentioned at the end of my essay back in August [20Aug25].

Or, the misplacement of the Turkey and Town Creek individuals might be attributable to a couple mislabeled sample tubes.  Or a spreadsheet error.  Any one of a million things could have gone wrong, who knows?  The larger point I am making here is that, because we bring an understanding of the evolutionary relationships of these animals to Nathan’s elaborate phylogenetic analysis, we can recognize error when it occurs.  Those two snails should have clustered with the conspecific Pistol Creek individual below.

In any case, regardless.  One might pray that, at the conclusion of his complex and intricate study, with his voluminous set of AHE results unfolded across the desk in front of him, showing the eight snails he had initially identified as “Elimia clavaeformis” splattered into four clusters across three quadrants of his colorful and richly-detailed Figure 4, Nathan Whelan might have been able to diagnose his prodigious clavaeformis confusion.  Alas, no. Ultimately, Nathan identified the incongruous collection of pleurocerids depicted in his Figure 5 as “two distinct lineages of E. clavaeformis and of its putative synonym, E. acutocarinata.”  Of such is prodigious polyphyly born.

I have said it, by actual count, 73 times over the 28-year history of this blog, but I am going to say it again, number 74, count ‘em all.  Gene trees are dependent variables, not independent variables.  They can be useful to test an hypothesis brought to them; they cannot yield an hypothesis to test.  They are the last chapter of your dissertation, not the first. Only if you have developed an understanding of the evolutionary relationships of a set of organisms by finer, independent means, can you interpret what a gene tree is telling you.

But let us finish this month’s essay on the same positive note with which we ended in September.  Bringing a finer understanding of the evolutionary relationships among the North American pleurocerid snails with us to the analysis, Nathan Whelan’s AHE gene tree really does have something important to tell us about the entire pleurocerid forest.  Which we will interpret together, next time.

Notes:

[1] Whelan, N. V., Johnson, P. D., Garner, J. T., Garrison, N. L., & Strong, E. E. (2022). Prodigious polyphyly in Pleuroceridae (Gastropoda: Cerithioidea). Bulletin of the Society of Systematic Biologists, 1(2). https://doi.org/10.18061/bssb.v1i2.8419

[2] My blog posts on the Pleurocera simplex/gabbiana sibling species pair:

• The cryptic Pleurocera of Maryville [13Sept16]
• The fat simplex of Maryville matches type [14Oct16]
• One Goodrich missed: the skinny simplex of Maryville is Pleurocera gabbiana [14Nov16]

[3] Well, I suppose I might fuss a bit about identifying any population of pleurocerids as Leptoxis “ampla.”  At least three good studies have been published in recent years strongly suggesting that Anthony’s (1855) nomen ampla is a junior synonym of Conrad’s (1834) picta.  More next month.

[4] I don’t have any insight to offer regarding the evolutionary relationships between pleurocerid populations identified as “Elimia crenatella” in Alabama and any of the other scores of nominal pleurocerid species inhabiting the malacologically benighted Mobile Basin.  This is because I have very little field experience in the region to draw upon.  Still waiting on my invitation, long time coming.

 

[5] For more about the fascinating evolutionary biology of Lithasia geniculata, see:

• Intrapopulation gene flow: Lithasia geniculata in the Duck River [7Dec21]
• What is character phase disequilibrium? [4Jan22]
• The third-most amazing research results ever published for the genetics of a freshwater gastropod population, and the fourth-most amazing, too. [3Mar22]

[6] Dillon, R. T. (2011) Robust shell phenotype is a local response to stream size in the genus Pleurocera (Rafinesque 1818). Malacologia 53: 265-277. [pdf]

 

[7] My series of published notes on the simplex/gabbiana sibling species pair:

• Dillon, R. T. (2016) Two reproductively isolated populations cryptic under Pleurocera simplex (Say, 1825) inhabiting Pistol Creek in Maryville, Tennessee.  Ellipsaria 18(2): 15-16. [pdf]
• Dillon, R. T. & J. D. Robinson (2016) The identity of the "fat simplex" population inhabiting Pistol Creek in Maryville, Tennessee.  Ellipsaria 18(2): 16-18. [pdf]
• Dillon, R. T. (2016)  Match of Pleurocera gabbiana (Lea, 1862) to populations cryptic under P. simplex (Say, 1825).  Ellipsaria 18(3): 10 - 12.  [pdf]

[8] Some not-insubstantial irony here.  From 2016 - 2020 I regularly published my technical results as notes in the FMCS Newsletter, “Ellipsaria.”  That stopped in 2020, when it was called to the attention of editor John Jenkinson that Ellipsaria contents were being indexed by Google Scholar, and hence that results such as mine might be reaching a wider audience.  According to the minutes of the FMCS Board Meeting of November 2, 2020, the complaint that prompted Dr. Jenkinson’s decision was lodged by one Dr. Nathan V. Whelan.

[9] I caught a small error in the caption of Nathan’s Figure 5.  Shells A and F were collected from Turkey Creek, not A and E.

 

[10] I both elaborated and belabored the genetic similarity between East Tennessee populations of Pleurocera clavaeformis and Alabama populations identified as “Pleurocera prasinata” back in 2016, here:

• Pleurocera clavaeformis in the Mobile Basin? [12July16]

[11] For a review of intraspecific hybridization in the North American Pleuroceridae, see:

• Widespread hybridization between Pleurocera laqueata and P. troostiana in streams of the Tennessee/Cumberland [15Oct24]
• Reticulate evolution in the North American Pleuroceridae [12Nov24]

[12] And for a review of the taxonomic consequences of widespread hybridization in pleurocerid populations referred to arachnoidea, edgariana, curreyana, and 23 other specific nomina, see:

• Taxonomy of the Pleurocera laqueata/troostiana complex. Part I [10Dec24]
• Taxonomy of the Pleurocera laqueata/troostiana complex, Part II [14Jan25]