Open pozsgaig opened 1 year ago
@pozsgaig thanks for pointing out these suspicious claims. Are you saying that there's no known Lepidopteran parasite of Spodoptera frugiperda ?
Also, when searching the rglobi for individual claims of Lepidopteran parasites of Spodoptera frugiperda via
first_round<-rglobi::get_interactions_by_taxa(sourcetaxon = "Spodoptera frugiperda", interactiontype = "hasParasite", targettaxon = "Lepidoptera", returnobservations = T)
I get:
> first_round[, c("source_taxon_path", "target_taxon_path", "study_citation", "study_source_citation")]
source_taxon_path
1 Animalia | Arthropoda | Insecta | Lepidoptera | Noctuoidea | Noctuidae | Spodoptera | Spodoptera frugiperda
2 Animalia | Arthropoda | Insecta | Lepidoptera | Noctuoidea | Noctuidae | Spodoptera | Spodoptera frugiperda
target_taxon_path
1 Animalia | Arthropoda | Insecta | Lepidoptera | Yponomeutoidea | Plutellidae | Plutella | Plutella xylostella
2 Animalia | Arthropoda | Insecta | Diptera | Tachinidae | Gonia
study_citation
1 Biever, K.D., Andrews, P.L. 1984. Susceptibility of lepidopterous larvae to Plutella xylostella nuclear polyhedrosis virus. J. Invertebr. Pathol. 44:117-119.
2 Arnaud, Paul Henri. A Host-parasite Catalog of North American Tachinidae (Diptera). Washington, D.C.: U.S. Dept. of Agriculture, Science and Education Administration, 1978.
study_source_citation
1 Sarah E Miller. 12/13/2016. Species associations manually extracted from Onstad, D.W. EDWIP: Ecological Database of the World's Insect Pathogens. Champaign, Illinois: Illinois Natural History Survey, [23/11/2016]. http://insectweb.inhs.uiuc.edu/Pathogens/EDWIP.
2 Sarah E Miller. 5/28/2015. Arnaud, Paul Henri. A Host-parasite Catalog of North American Tachinidae (Diptera). Washington, D.C.: U.S. Dept. of Agriculture, Science and Education Administration, 1978.
Strangely, this result includes fewer hits than your "grep" search. I need to look into this more to understand what is going on. Hoping to get back to this sooner rather than later. Apologies for the delay, and let me know if you find out some addition details.
@jhpoelen "Are you saying that there's no known Lepidopteran parasite of Spodoptera frugiperda ?" - Indeed, nothing like that exists. Only a few Lepidoptera are parasites but, according to my best knowledge, none of them parasitises other Lepidoptera. Also, the returned species are common agricultural pests, as is the source species, so the link between them is probably the common enemies (viruses, or other pathogens). I am also trying to have an in-depth look later this week.
@jhpoelen Jorrit, I have further investigated this issue and I found that querying some parasitic wasps also brings up plant species as being parasitised. These wasps are host specific and linked to Lepidoptera species, so this should not happen. When I searched the source, I found that, for example, Trichogramma sibiricum 'is parasite of' Vaccinium macrocarpon. This hit also comes up on the website, and I checked where does it originally come from. The record is from the NHMS's Universal Chalcidoidea Database (UCD) and for this particular wasp species that database lists several Lepidoptera as hosts and some plants (also bacteria) as well as ASSOCIATED SPECIES. Based on the linked article on the site (https://www.nhm.ac.uk/resources/research-curation/projects/chalcidoids/pdf_X/HaleEl2002.pdf), the wasps were used as biocontrol agents on cranberry (Vaccinium) plants against the Lepidoptera pest. Thus, it seems, that the interaction is erroneously recorded in GLOBI. If all interactions from UCD were automatically put down as parasitic relationships, this can be a rather large issue and impact thousands of records. I am, actually, not sure this has caused the discrepancies reported in this issue - but there is a fair chance. Gabor
@pozsgaig Thanks for looking into this, and I am glad you were able to trace the provenance of these suspicious claims. With these details, I am hoping to update the translations that GloBI applies to the provided NHMS's Universal Chalcidoidea Database (UCD) database.
@pozsgaig I've recorded your refutation record in https://github.com/globalbioticinteractions/refuted-interaction-data/commit/8ab40bafa2070900e55f9fe3b8a82cabc064ca25 as part of https://github.com/globalbioticinteractions/refuted-interaction-data . Once indexed, these records will show up in the "refuted by" section on the website. If you have more claims that you'd like to refute, please feel free to create a pull request, or even create your own refutation dataset. Your observations, and similarly, those of your colleagues are valuable insights that I think should be recorded and distributed.
That said, I am hoping to get to the details of GloBI's translation/indexing of the NHMS's Universal Chalcidoidea Database (UCD) database next.
Thanks for being patient.
@pozsgaig I was able to trace the claim to the its source and reported the issue at https://github.com/globalbioticinteractions/natural-history-museum-london-interactions-bank/issues/6#issuecomment-1503257496 . It appears that GloBI indexes the claim as stated, and that the claim itself may be in need of revision/ discussion.
When I search for parasites of the fall armyworm, other Lepidoptera species appear among the target taxa (e.g. Helicoverpa zea).
The problem may be with the viruses since there seems to be no standard of reporting their names and some automated taxon search may pick up the Lepidoptera taxonomy if the word "virus" is not clearly included in the name - just guessing though.