Which ants should we target for genome sequencing?

This weekend, Arizona State University is hosting a slate of myrmecologists to brainstorm on ant genomes.  I'd link to the meeting information, but apparently the gathering is so informal that they've not given the event a web page.  In any case, the topic is this:  in the age of (relatively) cheap genomes, which ants should we sequence? And, what should we do with the assembled data?

I originally planned to attend, but life intervenes and I'm frozen to the tundra of central Illinois.  Instead, I will register here a few suggestions about which species should considered, in addition to the already-funded projects (Harpegnathos, Camponotus, Solenopsis and Pheidole).  My criteria are twofold.  First, the ant must occupy a phylogenetic position that will maximise insight when considered with the exisiting genomes.  Second, the ant should have some additional property whose study will benefit from genomic information.  Here's the list:

1. Linepithema humile (the Argentine Ant) will likely be sequenced soon.  But as the funding is not yet secure, I'll push for Linepithema anyway as the most logical next genome.  Reasons to sequence the Argentine Ant abound.  It is a model species for nestmate recognition and for biological invasions, already employing dozens of scientists in many countries.  It causes considerable economic damage.  It will be the first ant sequenced from the great and speciose subfamily Dolichoderinae.  And it will also be cheap:  Linepithema humile has one of the smallest measured ant genomes.

2. Myrmecia pilosula (Jack-Jumper) is the most deadly Australian ant, owing to a sting with a particularly allergenic blend of enzymes.  But that's not the main reason to sequence Myrmecia.  The complex of species associated with M. pilosula has an astounding range of chromosome variants, from a highly fragmented 90 chromosomes to a species with the entire genome linked together, astonishingly, on a single giant chromosome.  A project here would be of great benefit to understanding the morphology of genomes, and how and why genomes are partioned into chromosomes.  Myrmecia are also large enough for physiological, neurological, and developmental research. Plus, as a member of the more recent formicoid clade with a relative simple social structure, they will serve as a nice control for teasing out which of the differences between the ASU project's Harpegnathos and Camponotus/Pheidole are due to social structure and which are merely phylogenetic carryovers.

3. Atta (Leafcutter Ant). No shortage of reasons to sequence Atta. The leafcutter ants, their fungi, and the constellation of associated microbes have become a leading model system for co-evolution of complex systems, and several laboratories are devoted to their study.  Atta also has an unusually well-developed caste system whose secrets could be answered through a bit of genomic sleuthing.  As a bonus, Atta keeps well in captivity.  That, and Atta is the most damaging agricultural pest in the new world tropics.

4. Paraponera clavata (Bullet Ant).  Not a pest, or a model species for anything, but an ant that should be included simply for the evolutionary insight of having a relictual early lineage to serve as a counterpoint for the more socially complex formicoids.  Plus, this ant is huge and hence suitable for the physiological/developmental spin-offs of a genome project.  As an alternative early lineage, I'd also consider Amblyopone.

5. Cataglyphis. A great deal has been learned about insect navigation in this desert ant.  A genome project would allow researchers a very powerful toolkit for probing at a molecular level how ants process information.

6. Pogonomyrmex barbatus (Harvester Ant). This insect has become something of a workhorse for desert ant research, hosting scores of studies on ecology, task allocation, and more recently, some funky genetics relating to hybridization and speciation.  A genome would assist all of these areas.  Plus, P. barbatus is a monomorphic myrmecine that can balance out the polymorphism of the other sequenced myrmicines, as well as spanning the root node for the subfamily.

7. Formica rufa-group species (Wood Ants). The emblematic Formica of the boreal northern hemisphere was the first ant named under Linnean taxonomy, and these ants continue to be subjects of interest in ecology, physiology, and behavior.  Formica are used in biological control of forest pests, and some species are among the very few ants protected by endangered species legislation.  The genus is an enormous group containing many degrees of social parasitism among its members, and a genome would help researchers gain a foothold into the molecular side of life history evolution.

8. Finally, we need desperately to sequence a non-ant.  Scientists are peeing themselves in the excitement over insect sociogenomics, but we need to keep perspective.  Everyone wants to do the sexy experiment but no one wants to run the boring control.  Our ability to say anything about the genetics of social behavior is greatly hindered by not having a nonsocial point of reference.  If all the aculeate genomes we have are social- and it looks now like we'll have Apis (social), Harpegnathos (social), Camponotus (social), Pheidole (social) and Solenopsis (social), then we still won't be able to finger the genes that drove the evolution of the insect societies.  Only through comparing the genomes of social and related non-social species will we be able to pick out the key differences between the groups.  So we should do Pepsis.  Or Sphex.  Or Dasymutilla.  It'll add value to all the other genomes.