NC State University Insect Museum

I am currently in Boston, MA (my hometown!) for the 2009 SICB annual meeting, where today I participated in the ontology workshop organized by Phenoscape. These ontology workshops are always inspirational for me, and, despite an unusually low turnout, this one was no exception. My head is swimming with ideas for our ontology project (the Hymenoptera Anatomy Ontology; check out the Hymenoptera Glossary (HAO), as well): policy changes, previously unknown references I need to read/cite, new collaborations to follow through on, new ways to employ the HAO in concert with other ontologies, alternative ways to describe hymenopteran phenotypes, etc. I'll be a Triangle Sabbatical Scholar at NESCent this semester, focusing on the use of ontologies in evolutionary biology; there will undoubtedly be some ontology-related posts in the coming months.

SICB is one of those scientific organizations that's been flying under my radar until now - I knew it existed but never paid much attention (too many other societies in my life!). I have to say, though, this meeting has some exciting talks on the docket. Here are seven I wish I could have seen but didn't:

• Deciding when to hatch: Predator and embryo cues in wasp-induced hatching of red-eyed treefrogs (HUGHEY, M.C.; ROGGE, J.R.; WARKENTIN, K.M.) - Did I read that right? Wow. It's amazing that Hymenoptera can have such a dramatic impact on Vertebrata. I guess I always think of anurans as the predators, rather than the prey.
• Performance of Shark Teeth during Puncture and Draw: Implications for the Mechanics of Cutting (WHITENACK, L.B.; MOTTA, P.J.) - OK, it's not entomology, but I do maintain an emerging interest in insect bite force and biomechanics and their applications in medicine (stay tuned). Besides, sharks are definitely amazing organisms.
• Diving wasps: swimming and flying at very low Reynolds numbers (MILLER, LA; SANTHANAKRISHNAN, A) - It's about fairyflies (Hymenoptera: Mymaridae)! This family includes the smallest known insect, Dicopomorpha echmepterygis, a parasitoid of bark lice (Psocodea). Just how do these teenie meenie insects fly?! They're <0.2 mm long!
• Recent and Ongoing Collections Facility Renovations at the Museum of Comparative Zoology (Harvard University) (BALDINGER, A.J.; CHUPASKO, J.M.; FORD, L.S.; HANKEN, J.; HARTEL, K.E.; ROSADO, J.; TRIMBLE, J.) - Maybe this will be us someday soon. I'd love to read about how they prioritized their improvements.
• Phylogeny and evolution of eusocial insects: a comparison of origins and losses in ants and bees (BRADY, S.G.; DANFORTH, B.N.; CARDINAL, S.) - My homeboys and girl talking about Hymenoptera.
• Direct visualization of hemolymph flow in the heart of a grasshopper (LEE, W.-K.; SOCHA, J.J.) - I am a sucker for talks/papers where synchrotrons are applied to morphological problems. I bet this talk was awesome. This will undoubtedly be us soon.
• Aerodynamics of Small Insect Flight and the Role of Bristled Wings (SANTHANAKRISHNAN, Arvind; MILLER, Laura ; DICKSON, William; DICKINSON, Michael) - I thought for sure this one would be about mymarids as well, but the subject here is thrips in the genus Haplothrips (Thysanoptera). I can't say that I think of thrips as being strong fliers, but I bet this talk has applications to many of my microhymenoptera with fringed wings.

I will certainly pay attention to this meeting's program next year. Great stuff abounds here.

Our dedicated undergraduate worker Caitlin completed the teaching collection rebuild awhile back and transitioned to a new, incredibly important initiative in the Insect Museum: databasing our specimens for collecting event data. Our admittedly lofty goals include assigning a unique identifier to every specimen in the museum (barcode label, ideally), grabbing the label data from each specimen (including each determination), imaging each specimen for multiple standard views, and presenting all of this information through the Web.

The research collection has already been inventoried for which species we have (and the number of specimens of each) by my predecessor, Lew Deitz. Thanks to his and Bob Blinn's efforts we know we have close to 30,000 species in the NC State University Insect Museum, which is pretty awesome. Even more impressive, to me anyway, is that we have a (conservatively) estimated 1.4 million specimens. Not too shabby for a university collection. Labeling, imaging, and capturing data from ALL of these specimens will take...ummm...awhile. Our test case, in the meantime, is the genus Bombus (Hymenoptera: Apidae) - the bumble bees. Data are just now rolling in, and all I can say is the results are exciting (as you will see in a couple months). We'll establish a separate series of Web pages when we complete the data entry for this lineage. Until then here are some NC collecting events for our American bumble bee (Bombus pensylvanicus (De Geer, 1773)) specimens, as viewed in Google Earth (or try it yourself, with our KML file - right-click, save as whatever.kml)

I saw that my colleague Johan Liljeblad's latest tome (reprint) came out today in Zootaxa. The manuscript covers the evolution of oak gall wasps (Hymenoptera: Cynipidae: Cynipini), which is cool in and of itself, but what I like is their use of standard image collections (including >2000 images) in Morphbank to illustrate scorings for 283 morphological characters. This allows "one-click access to information that is very close to the raw data on which [their] observations are based" (pg. 4). The authors also submitted their matrices (15062 and 15063) to TreeBASE. So why is all of this important? Well, it's an example of how the field or morphological phylogenetics is inching closer to science that's verifiable and repeatable. Take this character, for example:


How do I know what "very weak" means in state 0 versus "short" in state 1? Well, with the matrix in hand I can check the image collection and see how they coded the taxa. The next step is to create collections of collections in Morphbank (one for each matrix, for example; is it possible?) and to use annotation tools to create truly unambiguous character state scorings. Awesome stuff. Congrats Johan!

Norm Johnson et al. published a revision (reprint) of Heptascelio (Hymenoptera: Platygastridae) yesterday in Zootaxa that makes use of LSIDs for authors, collections, literature, specimens, taxon concepts, etc. Maybe this is the first case for LSIDs in insect taxonomy? I can feel the momentum building...

NB: Heptascelio are all parasitoids of grasshopper (Orthoptera: Acrididae) eggs (similar to Chromoteleia on Tettigoniidae), and adults are often phoretic on the host mom. Cool!

I am very excited to report that the Museum now has its very own LSID [link] thanks to the folks at TDWG (particularly Roger Hyam) as part of the Biological Collections Index. Great! What does that mean, and why is it cool? Lets say someone maintains a database of insect taxonomy, and archived within this conglomeration of data is information on the type specimens we house and protect here at NCSU. How does this database refer to our collection? As the North Carolina State University Insect Museum? The NC State University Insect Museum? By our coden, NCSU?

What if another database archives images of type specimens, and they refer to us the NC State University Insect Collection (which is part of the Museum)? And, to confuse this situation even more, what if we decide to change our name, thanks to the limitless generosity of the world's richest man, to the Dairy Queen® Museum of Insects?

Well, any of these scenarios (and synonyms) could potentially lead to miscommunication down the line - e.g., the taxonomy database calls on the image database to serve up photos of the 'NC State University Insect Museum' types, and the image database responds with a "ummm...we don't have any of those" message because they're all archived as being at 'NCSU'. When the data entry folks try to sort out the situation they see that no such entity exists in Raleigh, NC - only the curious (and delicious) DQ Insect Museum resides there.

By using globally unique identifiers (GUIDs; an LSID is a GUID) we can prevent these kinds of useless and frustrating interactions. Both databases (in fact all databases) would simply refer to us as urn:lsid:biocol.org:col:1024, so that they are assured that they point to the same collection of insects. Those databases would also, in an ideal world, use LSIDs for other objects, including taxon names, taxon concepts, references, publications, author names, etc. Check out examples at the LSID resolver. Here's how ours looks at the moment. Expect a more fully developed resource to be unveiled this summer. Sweet!

Rod Page (University of Glasgow) has an excellent paper in press that explains GUID utility and sructure. Very cool stuff.