Members of the Braconidae are mostly (99.9%) parasitoids of other insects and many are important in the natural and biological control of insect pests. Empirical research on the systematics of the Braconidae has centered on members of the subfamilies Agathidinae, Cenocoeliinae, and Sigalphinae. This includes revisionary taxonomy and phylogenetic analysis at the species level and at generic and higher taxonomic ranks.

    We are investigating methods for summarizing competing phylogenetic hypotheses. Sharkey et al. (2012) describes a weighted compromise method that corrects for a bias inherent in majority-rule consensus or compromise trees when the source trees exhibit non-independence due to ambiguity in terminal clades. We are currently investigating the usefulness of this method as an alternative for summarizing the tree files produced by Bayesian analyses.

    Sharkey, M.J., Stoelb, S.A.C., Miranda-Esquivel, D.R., Sharanowski, B.J. 2012. Weighted compromise trees: a method to summarize competing phylogenetic hypotheses. Cladistics. 29: 309-314.

    Back to Top

    Several taxa of neotropical Braconidae, representing hundreds of species, are part of a large mimetic complex that includes ichneumonid wasps and reduviid bugs. By examining phylogenetic patterns, we are testing the hypothesis that some of these taxa co-speciated and co-evolved mimetic patterns.
    Back to Top

    With the advent of next-generation sequencing, we have been conducting studies to determine the species-richness in arthropod samples that contain over 100,000 specimens and thousands of species. Traditional practices to obtain this information have cost millions of dollars, years of field and lab work, and employed large teams of taxonomists. Therefore repeating such studies for comparative purposes is not economically practical. With next-generation sequencing (e.g., Illumina MiSeq), it is possible to generate between 10-20 million sequences in a couple of days for ~ $2000. Using this technology it is possible to amplify a single gene from a sample of insects and in theory, have all of the species sequenced for a given gene. So for just a few thousand dollars, it is possible to accurately estimate the number of arthropod species in a natural area. This technique will provide land managers, faced with difficult decisions about which areas to preserve, with another valuable tool to determine which areas should receive protection.

    In a preliminary study, we pooled 36 Malaise trap samples from a site in south-central Kentucky, sieved them into 4 size fractions and extracted DNA from the fractions in bulk. We then amplified the cytochrome c oxidase subunit I gene using general primers and submitted the PCRs for sequencing on the Illumina MiSeq sequencing platform. Bioinformatics analyses of the sequences resulted in an estimate of 4,225 arthropod species among 308 families, 30 orders, and 6 classes. Of these, 607 matched an identifiable species at >98% sequence similarity on the BOLD database.

    On a larger scale, our ultimate goal is to combine the data from samples taken across the globe in order to gain a more accurate estimate of global biodiversity. Estimates between 3-100 million species have been proposed in recent decades. If the most recent estimates of approximately 6.1 million arthropod species prove accurate, then over 250 years of Linnaean-based taxonomy has only described about a fifth of the earth's biodiversity. The utilization of our procedures at sites across the globe could go a long way towards determining the number of species on earth. We have received a grant from the Kentucky Science and Engineering Foundation (KSEF-3068-RDE-017) to refine our techniques from samples taken in natural areas in Kentucky.

    Back to Top

    Investigating lowland Amazonian rainforest in Peru CENTRO DE INVESTIGACIONES ALLPAHUAYO
    Back to Top