Submitted by Benjamin Blonder on
Leaf venation architecture is highly variable among plant species (e.g. single, branching or reticulate veins) and across spatial scales (e.g. one or multiple vein orders). From an evolutionary perspective, some aspects of network architecture may be more variable/labile than others. For example, it has been suggested that some very large and very small scale patterns in venation network architecture reflect deep phylogenetic niche conservatism and/or limited functional variation, while intermediate-scale patterns may evolve rapidly and may reflect adaptive responses. Due to the difficulty of collecting network architectural data for whole-leaves, prior studies have mostly focused on categorical descriptors of spatial scale (higher vs. lower vein orders). Here, we will apply an innovative approach to fully describe multiscale network properties on whole-leaves of a phylogenetic broad set of plant species (ca 10.000 cleared leaf images publicly available at Smithsonian and UCMP databases). Then, we will investigate whether the phylogenetic conservatisms of vein architectural traits vary across spatial scales and whether there are detectable temporal trends in this architectural variation. Deciphering such evolutionary trends would advance theory for the development of transportation efficiency, not only in leaves, but also in other types of spatial networks (e.g. informational, traffic and electrical systems).
We will recruit up to two undergraduate students to specifically work on:
i.Hand-tracing cleared leaf images using image processing softwares (ImageJ/GIMP);
ii.Extracting multiscale network architecture using machine learning algorithms (MATLAB) in a high performance computer;
iii.Managing/organizing large datasets of species taxonomic hierarchy and architectural traits;
iv.Conducting phylogenetic analysis, e.g. temporal trends in vein architectural variation, test for phylogenetic conservatisms (K statistic under a Brownian evolution model) in vein architectural traits across spatial scales.
Students will develop skills in machine learning, imaging processing (ImageJ, GIMP) and computational data analysis with R/MATLAB. This will also be an excellent opportunity for students to improve their independent thinking skills and to learn how to access and conduct analysis in a high performance computer. Students will also be welcomed to engaged (remotely) in weekly lab meeting activities events organized by the Macrosystem Ecology Lab, e.g. research presentations, paper discussions, orientation sessions.
The project will be 100% remote unless public health conditions and restrictions change.
Candidates are required to be detailed oriented and to show a strong motivation to develop data management and data analysis skills. Biology major are preferred. Prior experience in processing images, managing large datasets or in data analysis with R is desirable, but not required. Sophomore, Junior or Senior are preferred.
Students with an interest in working with a diverse team (Brazil, Ghana, USA) are encouraged to apply. The lab provides an inclusive and supportive work culture. Current lab members are listed at http://benjaminblonder.org/people/. A $1000 stipend may be available for students who have overcome significant adversity to be present at the university.