Page 1 of 1

Complete mitochondrial genome of some random loricariine misidentified in a pet store

Posted: 01 Sep 2020, 18:19
by bekateen
Can anyone assist with the ID of the loricariine fish used in this study? It's cetainly not Sturisomatichthys panamensis. Sadly, only one dorsal-view photo is available from the paper. What disturbs me most about this paper is that although I can't ID the fish (my first thought is a Rineloricaria) the color pattern of the specimen shown has no similarity to any Sturisomatichthys at all (based on the photos we have here). And yet, the paper's COX1 sexquence nests this specimen so nicely inside Sturisomatichthys, presumably along side other samples reported as "Sturisoma panamense" by past authors (is it be from the same lab?).

If this fish is Sturisomatichthys panamensis (or any Sturisomatichthys for that matter), it's unlike any photo I've seen. And if it's another genus,... and if the COX1 sequence is correct, it should be unsettling for the scientific interpretation of COX1 sequences and their meaning for taxonomy.
Ugh...


Ren, F., Chen, D. & Ma, X. (2020). The complete mitochondrial genome of Sturisomatichthys panamense (Siluriformes: Loricariidae) analysed by next-generation sequencing and phylogeny of the catfish subfamily Loricariinae (Siluriformes: Loricariidae). Biologia, 75, 1365–1372. https://doi.org/10.2478/s11756-019-00389-7

https://link.springer.com/article/10.24 ... 19-00389-7
PDF: https://link.springer.com/content/pdf/1 ... 0389-7.pdf

Currently valid name:
Ren et al. (2020) wrote:ABSTRACT
(Loricariidae, armored catfishes) is widespread throughout Central and South America from Pacific slope rivers of Panama to Caribbean slope rivers of Colombia. Due to the incredible body shape, it becomes one of the most popular pet fish in the world. However, little research has been done for this species, especially its genetic resource. In the present study, the complete mitochondrial DNA genome of S. panamense was first reported by next-generation sequencing method. The entire length of mitochondrial genome is 16,497 bp and the nucleotide composition was made up of 31.2% A, 28.3% C, 25.5% T, and 15.0% G, respectively, indicating an A + T (56.7%)-rich feature in the S. panamense complete mitogenome. The mitogenome includes 22 transfer RNA genes, 2 ribosomal RNA genes, 13 protein-coding genes and one control region. With the exception of 8 tRNA genes and NADH6, mitochondrial genes are encoded on the heavy strand, which was in line with other vertebrates. Phylogenetic trees centred on the Loricariinae were reconstructed using another 72 reported cytochrome c oxidase I (cox1) gene sequences. They indicated monophyly of the Loricariinae and presented a basal split within the Loricariinae resulting in two major tribes, the Loricariini and the Harttiini. Sturisomatichthys panamense was placed with and formed a sister group to in the phylogenetic trees.

Re: Complete mitochondrial genome of some random loricariine misidentified in a pet store

Posted: 01 Sep 2020, 19:52
by Jools
I'd go for but only on a hunch as opposed to any scientific reason. Seems to be in keeping.

Jools

Re: Complete mitochondrial genome of some random loricariine misidentified in a pet store

Posted: 01 Sep 2020, 20:00
by bekateen
Jools wrote: 01 Sep 2020, 19:52I'd go for but only on a hunch as opposed to any scientific reason. Seems to be in keeping.
R. lanceolata was my thought too, but there are so many Rineloricaria to choose from, if you allow for oddballs sneaking in from occasional imports. But most disturbingly, if their fish is ANY Rineloricaria, why is their COX1 tree throwing it in with all the Sturisomatichthys and Sturisoma. It's either really poorly done molecular work, poorly chosen computer settings for performing the tree building, or it shows COX1 genes are utterly unreliable as a taxonomic tree builder.

As far as the hobby goes, sadly, we almost never see R. lanceolata available around me. An occasional , but besides that, only R. eigenmanni. NationWide Aquatics had some R. lanceolata last year. The most novel thing on the scene in the USA right now is at Freshwater Exotics, unintentionally imported when the FWE thought they were getting a Xingu species of lip brooder.

Cheers, Eric

Re: Complete mitochondrial genome of some random loricariine misidentified in a pet store

Posted: 02 Sep 2020, 11:20
by dw1305
Hi all,
bekateen wrote: 01 Sep 2020, 20:00....... But most disturbingly, if their fish is ANY Rineloricaria, why is their COX1 tree throwing it in with all the Sturisomatichthys and Sturisoma. It's either really poorly done molecular work, poorly chosen computer settings for performing the tree building, or it shows COX1 genes are utterly unreliable as a taxonomic tree builder........
It looks like <"it is another one for ">, @racoll's", <"Spreadsheet of shame">.

cheers Darrel

Re: Complete mitochondrial genome of some random loricariine misidentified in a pet store

Posted: 02 Sep 2020, 14:57
by Silurus
Presumably one could check the Genbank accession numbers of the comparative sequences to see if those have been correctly identified.

Re: Complete mitochondrial genome of some random loricariine misidentified in a pet store

Posted: 02 Sep 2020, 17:53
by racoll
dw1305 wrote:It looks like <"it is another one for ">, @racoll's", <"Spreadsheet of shame">.
:)
Silurus wrote:Presumably one could check the Genbank accession numbers of the comparative sequences to see if those have been correctly identified.
They look legit. I checked COI and 12S. Wild caught vouchers from Panama.

I imagine what happened, was they went to a petshop and bought a crap ton of fish, then took photos of them and sequenced them. They then used the COI sequence to ID their fish, because they didn't know what they were. The closest was S. panamense (it's actually a lot closer to S. leightoni if you dig a bit deeper), so they wrote the paper up as that. Then, they probably just used the wrong photo for the figure.
bekateen wrote:But most disturbingly, if their fish is ANY Rineloricaria, why is their COX1 tree throwing it in with all the Sturisomatichthys and Sturisoma. It's either really poorly done molecular work, poorly chosen computer settings for performing the tree building, or it shows COX1 genes are utterly unreliable as a taxonomic tree builder.
As the saying goes, "never ascribe to malice that which can be explained by incompetence".

Re: Complete mitochondrial genome of some random loricariine misidentified in a pet store

Posted: 02 Sep 2020, 20:11
by dw1305
Hi all,
Now that is interesting and a pleasant surprise, I also suppose it makes a lot more sense that the image is the issue, rather than the gene sequence.

cheers Darrel