Work Package 1. Data exploration (Months 1-2)
I undertook a through survey on the fly wing dataset to identify outliers, potential ecological/life history covariates and to understand data structure.
Work Package 2. Investigate phylogenetic structure of evolvability (Months 2-6)
I measured mean evolvability, the ability of a trait to respond to a unit directional selection, of wing shape using the theory proposed by Hansen and Houle (2008). Time-calibrated molecular phylogeny including all 111 studied species is obtained from van der Linde et al (2010). I mapped the evolution of mean evolvability of female wing shape on Drosophilids phylogeny (Figure 2). Blue and red colors represent high and low evolvability in wing shape respectively, and branches were colored with gradient between these colors assuming Brownian motion model of evolution. I identified two lineages that have relatively high evolvability in wing shape: a clade of Hawai'ian flies and the Drosophila melanogaster subgroup. Male data showed the same pattern.
My result indicates that Hawai'ian subgroup and melanogaster subgroup have evolved higher potential for evolution in wing morphology than other species within the family. Hawai'ian subgroup is a remarkable case of island adaptive radiation with an estimated 1,000 species all arising from a common ancestor in the last ~25 millions of years (O'Grady et al 2011). Furthermore, the subgroup outstands by its exceptionally large body size compared to other members of the family (Figure 2). Thus, it might be the case that a high evolvability in wing shape is related to adaptive radiation. The melanogaster subgroup includes cosmopolitan human commensals (D. melanogaster and D. simulans) as well as island endemic species (D. sechellia and D. mauritiana). It may indicate that a high standing genetic variation in the melanogaster subgroup might enabled the cosmopolitan distribution and human commensal ecology of D. melanogaster and D. simulans. I am currently writing a manuscript based on these findings.
Work package 3. Developing the rate-trait model (Months 3, 7-9)
With my host Prof. Hansen at University of Oslo and a mentor Dr. Geir Bolstad at Norwegian Institute for Nature Research, I developed a new statistical model that relates trait(s) with the rate of evolution. I discussed with Prof. Hansen and Dr. Bolstad over potential evolutionary models suitable for this method in June, in Trondheim. We further exchanged ideas in October, in Oslo. We decided that a model that assumes a trait behaves as the Brownian motion (BM) and the rate is related to this BM trait. This model is the simplest evolutionary model among the models that we considered. Further consideration on model properties of this simplest model is necessary to build more complex models, where the trait behaves as more complex evolutionary models.
I and Dr. Bolstad are currently developing a program code in R statistical environment. In addition, to test the rate-trait model upon completion of program writing, I collected the data of brain size and beak morphology from approximately 2,000 species of birds. I will use this data to i) examine the details of model properties, ii) identify and fix computational problems in running the rate-trait model, and iii) test the Wyles' hypothesis (Wyles et al. 1983) stating that large brains are related to increased rate of morphological evolution. Our progress is still preliminary. However, I am expecting to generate at least two papers in a near future; i) description of the rate-trait model and ii) a test for Wyle's hypothesis.
References
Hansen F.T. and Houle D. 2004. in Phenotypic integration: studying the ecology and evolution of complex phenotypes, ed. Pigliucci, M. and K. Preston, Oxford University Press, Oxford. pp. 130-150.
Hansen F.T. and Houle D. 2008. J. Evol. Biol. 21:1201-1219.
Houle D. et al. 2003. BMC Evol. Biol. 3:25.
ven der Linde K. et al. 2010 Genet. Res. 92:25-38.
O'Grady P.M. et al.2011. Mol. Phylogenet. Ecol. 58:244-256.
Steppan S. et al. 2002. Trends Evol. Evol. 17:320-327
Wyles et al., 1983. PNAS. 80: 4394-4397
