Student Research - Biology
Paige Barlow
Phylogeny of the Genus Breviceps Based on the Rhodopsin Gene
Abstract: Phylogenetic relationships of microhylid frogs are poorly understood. The first comprehensive molecular-based phylogenetic analysis for African microhylid genera was recently presented (Loader et al., 2004). This study sampled all subfamilies and six of the eight genera found in continental Africa. However, there were only two samples of Breviceps, a genus found in arid to semiarid areas of southern Africa, included in the analysis, both of Breviceps mossambicus (Frost, 2004). Therefore, the goal of this project is to assess a hypothesis of evolutionary relationships for Breviceps species based on the nuclear gene Rhodopsin 1 (Rhod1). Tissue samples from 13 individuals representing 9 of 15 currently recognized species in the genus were obtained. DNA was isolated, amplified, purified, and sequenced using standard protocols (Hillis et al., 1996). Sequence data for a fragment of approximately 350 base pairs (bps) of Rhod1 were collected. Probreviceps served as the outgroup in the analysis. Herein we present a hypothesis of evolutionary relationships for the species of Breviceps.
The Maximum Parsimony (MP) analysis resulted in 10 equally parsimonious trees of 79 steps with a consistency index of 0.911. Maximum Likelihood (ML) analysis was run under the General Time Reversible (GTR) model of DNA substitutions with gamma (G) distribution. Both analyses (MP and ML) resulted in very similar relationships. However, ML provided slightly higher resolution of some clades. MP recovered an unresolved group of species consisting of B. acutirostris, B. fuscus, B. gibbosus, and B. montanus (Group A). The same group was recovered under ML except that in the later, B. montanus has a basal position to the three other taxa. B. verrucosus was recovered as the sister species to Group A in both MP and ML. The species B. adspersus, B. bagginsi, and one sample of B. mossambicus, appear closer related to the [Group A-B. verrucosus] clade than to other species in both analyses. Furthermore, the two samples of B. bagginsi clustered together in both analyses. Whereas relationships of B. adspersus, B. bagginsi, and the B. mossambicus sample to the [Group A-B. verrucosus] clade are unresolved under MP, the B. mossambicus sample appears closer related to the [Group A-B. verrucosus] clade than the other species under ML. A second group of species with unresolved relationships among themselves (Group B) was recovered in both analyses. Species in this group are B. poweri and two samples of B. mossambicus.
The preliminary data suggest that B. verrucosus and B. bagginsi, species for which we have more than one sample, are valid species. However, for the other such species, B. mossambicus, the different samples do not cluster together. Rather, in all analyses, they appear to have relationships in highly divergent clades. It appears that B. mossambicus, as currently defined, may consist of more than one species.
Click here to view Paige's poster.
| Presenter |
Paige Barlow |
| Mentor |
Dr. Rafael De Sa |
|
