DNA Barcoding of Pygmy Hoppers—The First Comprehensive Overview of the BOLD Systems’ Data Shows Promise for Species Identification
Round 1
Reviewer 1 Report
1. I think authors should give a brief introduction for the systematic position of Tetrigidae in Orthoptera.
2. the gene name should be italic.
3. It has been shown in many studies that a single molecular marker such as COI is often difficult to clearly reflect the phylogenetic relationship of higher rank taxa, but a good gene for valuing the differences at specific level. so why authors perform a test on this gene for analyze high rank taxa.
Author Response
Dear Reviewer,
Thank you for your comments. Following are our point-by-point responses.
„1. I think authors should give a brief introduction for the systematic position of Tetrigidae in Orthoptera.“
We agree that this is important information and have added it to the Introduction.
„2. the gene name should be italic.“
Thank you, the gene name has been italicized.
„3. It has been shown in many studies that a single molecular marker such as COI is often difficult to clearly reflect the phylogenetic relationship of higher rank taxa, but a good gene for valuing the differences at specific level. so why authors perform a test on this gene for analyze high rank taxa.“
In some taxa, COI does maintain enough phylogenetic signal to be useful at higher taxonomic ranks, so our goal was to explore this possibility in Tetrigidae. However, these cases are rare and our results conform to the expectations of a weak phylogenetic signal. We believe that it is still important to report this finding, but agree that it is not necessary to go into such detail. We have thus dramatically reduced the amount of text discussing this issue.
Kind regards,
The authors
Reviewer 2 Report
An interesting contribution to the value of the currently known state of DNA barcoding information for Tetrigidae. The English text is correct and easily readable. There is only a single sentence that should be corrected.
Lines 315-317: incomplete sentence: These numbers pale in comparison to the global diversity of Tetrigidae (Cigliano et al. 2023)[15] and represent less than 5% of the currently described taxa.
should be either: These numbers are pale in comparison to the global diversity of Tetrigidae (Cigliano et al. 2023)[15] and represent less than 5% of the currently described taxa.
or: These numbers, pale in comparison to the global diversity of Tetrigidae (Cigliano et al. 2023)[15], represent less than 5% of the currently described taxa.
Author Response
Dear Reviewer,
Thank you for your comments.
In the marked sentence, we use the word „pale“ as verb, defined by the Collins dictionary as: „If one thing pales in comparison with another, it is made to seem much less important, serious, or good by it.“
With an example sentence: „When someone you love has a life-threatening illness, everything else pales in comparison.“
Therefore, we believe that our usage of the term is correct and will maintain it.
We appreciate your positive stance on our work and wish you all the best.
Kind regards,
The authors
Reviewer 3 Report
Referee’s comments:
In References authors must again check list carefully and made corrections.
1) Unify Volumes and Numbers of journals, because different citations presents. For example,
Hawlitschek, O., Morinière, J., Lehmann, G.U.C., Lehmann, A.W., Kropf, M., Dunz, A., Glaw, F., Detcharoen, M., Schmidt, S., Hausmann, A., Szucsich, N.U., Caetano-Wyler, S.A. and Haszprunar, G. DNA barcoding of crickets, katydids and grasshoppers (Orthoptera) from Central Europe with focus on Austria, Germany and Switzerland. Mol. Ecol. Resour. 2017, 17, 1037–1053.
Devriese, H., Husemann, M. Afrosystolederus garmsi (Orthoptera, Tetrigidae), a new genus and species from Mount Gibi (Liberia) with remarks on Systolederus, Pseudosystolederus and Teredorus. Zootaxa 2023, 5258(3), 331–341.
2) Add DOI numbers in some papers. For example,
Fernandez-Triana, J. L. Turbo taxonomy approaches: lessons from the past and recommendations for the future based on the experience with Braconidae (Hymenoptera) parasitoid wasps. ZooKeys 2022, 1087(4) , 199–220. https://doi.org/10.3897/zookeys.1087.76720
Note: the number of journal issue is missed!
or
Devriese, H., Husemann, M. Afrosystolederus garmsi (Orthoptera, Tetrigidae), a new genus and species from Mount Gibi (Liberia) with remarks on Systolederus, Pseudosystolederus and Teredorus. Zootaxa 2023, 5258(3), 331–341. https://doi.org/10.11646/zootaxa.5258.3.6
Finally, in Conclusion authors stress that “On the other hand, the COI gene does not carry a sufficiently strong phylogenetic signal that would allow for a simple reconstruction of relationships above the specific level. Phylogeny should be attempted with additional genes, including nuclear genes or even entire genomes”. But the system of Tetrigidae at subfamilies, tribes and even genera level is based on very old features proposed by Bolivar and Hanckock. This system is useful for determination of taxa but seems not to be reflecting the real phylogenetic relationships in this family. So, both factors (needs of studies of nuclear genes or even entire genomes and needs of reclassification of family) may be discussed.
Quality of English Language is good for understanding.
Author Response
Dear Reviewer,
Thank you for your comments. We have checked the references again and made corrections, thanks for pointing that out.
Regarding the comment on the conclusions, we of course agree that the current system does not reflect the true evolutionary relationships between taxa. In this paper, we do not delve deep into the taxonomy, but we have added a brief reflection on this issue in the Discussion section.
Kind regards,
The authors
Reviewer 4 Report
Abstract
Page 1, line 16: “…… species that are difficult to identify”, this problem needs to be stated in more detail.
Page 1, line 24-25: “The deeper nodes in the phylogenetic trees are not well-supported, indicating that this gene has a very weak phylogenetic signal beyond the specific level.” The conclusion cannot be supported by the results.
Introduction
Page 2, line 71: You need to clearly state how many subfamilies, genera and species are included in Tetrigidae, this involves the sample size.
Page 2, line 74-76: “The identification of species is difficult as most Tetrigidae are small, with many similar species for which no reliable identification keys exist; further, the group undergoes constant shifts in taxonomy”, these should not be a key problem in classification.
Page 2, line 83: “Therfore, ……”, spelling mistake.
Materials and Methods
Page 3, line 103: “96 sequences (4% of the data) were labeled as likely containing indels and were removed”, why remove these sequences?
Page 3, line 107-109: “…… only the sequences of records containing photographs were considered. The specimens were identified by comparing them to the original descriptions and the holotypes available on the Orthoptera Species File Website”, This means that the author didn't see the specimen and identified them from the photograph? How to ensure the accuracy of identification?
Page 3, line 113: “The final dataset of 183 COI sequences”, How many genera and species do these 183 sequences represent?
Page 3, line 132: “We used the GTR+G+I substitution model ……”. The method of model selection needs to be explained in detail.
Results
The author stated many problems of BOLD system and grasshopper classification, but the author did not solve these problems before the analysis. How to ensure the accuracy of the analysis and how to trust the results of the analysis.
Page 4, line 178: “The remaining genera, both within Tetriginae and in other subfamilies, are represented by relatively few records”. How to rule out that the results of phylogenetic analysis are not due to sampling problems?
Page 7, line 201: “Identification methods”. It is not ‘identification’, but ‘molecular species delimitation’.
Discussion
Page 12, line 296-297: “The phylogenetic analysis showed that COI barcodes show promise for the identification of most of the included species, but some clusters remain ambiguous”. Phylogenetic analysis is used to explore phylogenetic relationships, not to identify species.
Page 12, line 298-299: “The deeper nodes in the phylogenetic trees are not well-supported, indicating that this gene has a very weak phylogenetic signal beyond the specific level.” The conclusion cannot be supported by the results.
Figures 1-2 should be improved.
Author Response
Dear Reviewer,
Thank you for your detailed comments. Following are our point-by-point responses.
„Page 1, line 16: “…… species that are difficult to identify”, this problem needs to be stated in more detail.“ AND „Page 2, line 74-76: “The identification of species is difficult as most Tetrigidae are small, with many similar species for which no reliable identification keys exist; further, the group undergoes constant shifts in taxonomy”, these should not be a key problem in classification.“
These points have been expanded upon in the Introduction, with multiple different references. The added section is as follows:
The identification of Tetrigidae species is difficult due to several factors. Firstly, Tetrigidae exhibit polymorphism in coloration, body size, and wing length, making those characters difficult to use in isolation (Ahnesjö & Forsman 2003, Steenman et al. 2015, Zhao et al. 2016)[19–21]. Secondly, the classification and taxonomy of many groups within Tetrigidae is problematic, with many revisions conducted recently and with more still to come (Tumbrinck 2014, Tan et al. 2019)[18,22]. Continuing on the previous point, there is a lack of clear diagnoses for many of the described species and higher taxonomic categories (Tan et al. 2019)[22], which leads to the lack of identification keys for many taxa (Itrac-Bruneau & Doucet 2022)[23]. Finally, reference databases such as the OSF (Cigliano et al. 2023)[16] sometimes contain multiple different species under a single species (due to unresolved taxonomy or earlier misidentifications), which means that it is often necessary to examine the type specimen to determine relevant diagnostic characters and arrive at a confident identification (Tumbrinck 2019, Kasalo et al. 2023)[24,25]
„Page 1, line 24-25: “The deeper nodes in the phylogenetic trees are not well-supported, indicating that this gene has a very weak phylogenetic signal beyond the specific level.” The conclusion cannot be supported by the results.“
Besides the phylogenetic analyses, we performed the test of substitution saturation that shows that the sequences used in the analysis contain so many substitutions that back-mutations are likely and hence the phylogenetic signal is not reliable at the deeper nodes. The reasoning for the use of phylogenetic analyses is discussed in one of the responses below. In addition to the substitution saturation results, we cite several papers that indicate that COI by itself very often does not possess a strong enough phylogenetic signal beyond the specific level, which has been checked by substitution saturation tests on multiple different occasions. The other reviewers and the editors agree that the results are sufficient (and indeed over-discussed) to claim that COI has a weak phylogenetic signal beyond the species level, so hence we remain by our conclusions. According to their comments, the discussion of this issue has been shortened as this is a well-known phenomenon.
„Page 2, line 71: You need to clearly state how many subfamilies, genera and species are included in Tetrigidae, this involves the sample size.“
The Introduction has been expanded to elaborate more on the position of Tetrigidae within Orthoptera and the composition and diversity of Tetrigidae.
„Page 2, line 83: “Therfore, ……”, spelling mistake.“
Thank you. The mistake has been corrected.
„Page 3, line 103: “96 sequences (4% of the data) were labeled as likely containing indels and were removed”, why remove these sequences?“
Thank you for pointing out this unclearly written sentence. We have expanded it to indicate that we removed them to minimize the risk of technical errors or the inclusion of pseudogenes. It has been demonstrated, e.g. by Nugent et al. 2020 (https://doi.org/10.1139/gen-2019-0206) that even when there are no stop codons in the sequence, it can still be a pseudogene. Further, during sequence submission, BOLD itself tries to flag the sequences with technical errors (10.1111/j.1471-8286.2006.01678.x), but obviously, both it and the coil package are not 100% accurate. By performing this analysis, we demonstrate that both algorithms perform similarly, and by excluding the dubious sequences we remove the possibility of an error that could wrongly invalidate some of the species clusters.
„Page 3, line 107-109: “…… only the sequences of records containing photographs were considered. The specimens were identified by comparing them to the original descriptions and the holotypes available on the Orthoptera Species File Website”, This means that the author didn't see the specimen and identified them from the photograph? How to ensure the accuracy of identification?“
This section has been expanded to be clearer about the terms under which the specimens were considered identifiable. The Introduction has been expanded to be more clear about why Tetrigidae are difficult to identify and under which cases they can be identified, that being the presence of defined diagnostic characters. We identified only those specimens for which the diagnostic characters were known and clearly visible, the same thing that is done with physical specimens. The number of candidate taxa is also greatly narrowed down by the location data that is provided for each specimen, further easing the identification. The specimens that could not be confirmed were not considered so that they would not impact the analysis. This approach allows the analyzed dataset to be of high accuracy with respect to identifications. Obtaining the physical specimens of the entire dataset is simply impossible as it is impossible to get a loan from every institution holding the specimens, and there are many different institutions that have deposited their data to BOLD. Further, identifying species from photographs is a usual and well-supported practice, as indicated by many recent papers that use citizen science data for research (https://doi.org/10.3160/0038-3872-118.1.58; https://doi.org/10.1071/WR20154; https://doi.org/10.1371/journal.pone.0226534; https://doi.org/10.1371/journal.pone.0218614). The normal practice is to exclude the data that cannot be confirmed and include the records that clearly display the properties required for identification. All of us have published extensively on Tetrigidae taxonomy and are well-acquainted with their diagnostic characters.
„Page 3, line 113: “The final dataset of 183 COI sequences”, How many genera and species do these 183 sequences represent?“
The counts have been added.
„Page 3, line 132: “We used the GTR+G+I substitution model ……”. The method of model selection needs to be explained in detail.“
Thank you for spotting this. We determined the appropriate model for both analyses by use of the ModelFinder function in IQ-TREE. The implementation of parameters is a little different in MrBayes so the most closely related model was used instead. We have expanded the section on model selection to make this clear. The implementation of the ModelFinder function, however, is beyond the scope of this paper and is explained in the provided reference.
„The author stated many problems of BOLD system and grasshopper classification, but the author did not solve these problems before the analysis. How to ensure the accuracy of the analysis and how to trust the results of the analysis.“
The problems with the taxonomy of Tetrigidae are numerous and cannot be simply resolved since they require extensive taxonomic work that will take many more years. We have approached this problem by identifying only those species for which we can be confident that they can be identified. This is the reason why some species are unidentified and why the final species count is not a firm number. All analyses were performed only as far as the relevant data could be confirmed, as indicated in Materials and Methods section.
On the other hand, the problems of BOLD Systems in the context of Tetrigidae is not well known; it was one of the goals of this study to identify those problems and to report them to that they can be fixed by the administrators of the database. We are not affiliated with BOLD Systems so we cannot update the taxonomic backbone and we cannot get access to its private systems.
„Page 4, line 178: “The remaining genera, both within Tetriginae and in other subfamilies, are represented by relatively few records”. How to rule out that the results of phylogenetic analysis are not due to sampling problems?“
This issue was discussed in an earlier comment. Since we are dealing with COI, which is confirmed to have a weak phylogenetic signal, and since our other analyses support this, it is likely that the sampling would have no significant effect on the topologies of the trees since the deeper nodes carry no information and are not considered in the interpretation. Please also see the comment dealing with our usage of phylogenetic analyses where we explain that the inference of phylogeny is not the primary goal and a difference in sampling would only provide more species clusters to examine, i.e. the conclusions would remain the same. In the future, a more comprehensive sampling would be very welcome and will indeed be necessary if we aim to conclusively resolve the Tetrigidae phylogenetic tree, but the primary goal of this study is to examine the usefulness of barcoding in Tetrigidae, not their phylogeny.
„Page 7, line 201: “Identification methods”. It is not ‘identification’, but ‘molecular species delimitation’.“
While it is true that the BIN system and similar tools such as ABGD are used for molecular species delimitation, i.e. finding clusters of similar sequences that are intended to correspond to some taxonomic rank, usually species, in this section we indeed deal with identification methods. „Identification methods“ is a section of metadata attached to every record in BOLD, and is aptly named as it contains the information on what method was used to get the ID that is attributed to the sequence. The „BIN Taxonomy“ identification method refers to the instances when a user uploads a sequence and requests an identification, to which the BOLD website responds with a BIN code that contains sequences most similar to the one uploaded. Hence, in this context, It is clearly an „identification method“ and is referred to as such by the author of the system (https://doi.org/10.1371/journal.pone.0066213). In this section, we did not perform our own molecular species delimitation. We appreciate the valid point that was raised here, but our use of the terminology is in line with the information that we are trying to relay and will thus leave the section unaltered.
„Page 12, line 296-297: “The phylogenetic analysis showed that COI barcodes show promise for the identification of most of the included species, but some clusters remain ambiguous”. Phylogenetic analysis is used to explore phylogenetic relationships, not to identify species.“
We used the phylogenetic analysis not to identify species, but to test whether the sequences belonging to the same species cluster together. The results show that this is indeed the case and that these clusters largely correspond to clustering algorithms. The topology of the trees beyond these clusters is irrelevant, as is demonstrated by both the literature and our results, namely the substitution saturation test. We have expanded this sentence and added an explanation to Materials and Methods section to more clearly state that the purpose of the phylogenetic analysis is neither identification nor phylogeny, but a test of the resolution of COI at the level of species. Despite the name, phylogenetic methods can be used to test the clusters like in our example, and do not necessarily have to be interpreted as showing evolutionary relationships at deeper nodes (https://doi.org/10.3389/fevo.2019.00302). The question of how identification relates to barcodes was addressed in the response above. Thank you for noting the ambiguity in the quoted sentence.
„Figures 1-2 should be improved.“
These two figures show the composition of subfamilies and genera in the dataset by simple counts. As far as we can see, all the numbers and text are readable and all the necessary information is condensed in the figures without taking too many pages. Since other reviewers and the editors did not ask for corrections, we prefer the figures to remain as they are unless concrete concerns are raised at a later time.
Finally, thank you once again for engaging with our manuscript and providing extensive comments.
Kind regards,
The authors
Round 2
Reviewer 4 Report
We know that COI has a weak phylogenetic signal beyond the species level, but we don't think the method you use and the results you get from the phylogenetic trees are sufficient to explain the problem.
Author Response
Dear Reviewer,
Thank you for your response. We agree that it would take a larger and more comprehensively sampled dataset to conclusively and beyond any doubt prove that COI in Tetrigidae by no means possesses a phylogenetic signal beyond the level of species. We have removed the parts of the text discussing this issue on Tetrigidae subfamilies and have left only a brief overview of the literature that outlines the current knowledge on COI and its use in phylogeny. Our results correspond to what is expected in case of a weak phylogenetic signal, but we do not claim to have proved it, only that it is a likely explanation. We have also added notes on the need for further studies more expansive that should elucidate this issue and provide definitive proof.
Thank you once again for your detailed review.
Sincerely,
The Authors
