August 24, 2017
A $2.5 million National Science Foundation grant will launch oVert, a new initiative to "teleport" museum specimens from their shelves to the internet by CT scanning 20,000 vertebrates and making these data-rich, 3-D images available to researchers, educators, students and the public.
The University of Michigan is one of 16 research institutions included in the grant.
oVert, short for openVertebrate, complements other NSF-sponsored museum digitization efforts, such as iDigBio, by adding a crucial component that has been difficult to capture — the internal anatomy of specimens.
"The high-resolution, 3-D anatomical data provide the opportunity to compare morphological diversity without damaging specimens. This opens the doors to rare and endangered species that are often difficult to study," said biologist Priscilla Tucker, one of four U-M researchers involved in the project.
U-M is one of five designated scanning centers for the project and will contribute scans from more than 3,500 of its Museum of Zoology specimens — fish, amphibians, reptiles, birds and mammals. CT scanners at the Medical School and School of Dentistry will be used.
With virtual access to specimens through the oVert project, researchers could peel away the skin of a passenger pigeon to glimpse its circulatory system; a class of third-graders could determine a copperhead's last meal; undergraduate students could 3-D print and compare skulls across a range of frog species; and a veterinarian could plan a surgery on a giraffe at a zoo.
The project will encompass representative specimens from more than 80 percent of existing vertebrate genera, and a selection of these will also be scanned with contrast-enhancing stains to characterize soft tissues.
Above, a CT scan of a Pacific tailed frog. The "tail" is actually the male genitalia. Below, a CT of a hognose snake showing its last two meals: a salamander and a toad. (Photos by Ed Stanley, Florida Museum of Natural History)
"Our goal is to provide data that offer a foothold into vertebrate anatomy across the Tree of Life," said David Blackburn, oVert's lead principal investigator and associate curator of amphibians and reptiles at the Florida Museum of Natural History at the University of Florida, the grant's lead institution.
"This is a unique opportunity for museums to have a pretty big reach in terms of the audience that interacts with their collections. We believe oVert will be a transformative project for research and education related to vertebrate biology."
CT scanning is a nondestructive technology that bombards a specimen with X-rays from every angle, creating thousands of snapshots that a computer stitches together into a detailed 3-D visual replica that can be virtually dissected, layer by layer, to expose cross-sections and internal structures. The scans allow scientists to view a specimen inside and out — its skeleton, muscles, internal organs, parasites, even its stomach contents — without touching a scalpel.
As part of oVert, these images will be housed in MorphoSource, a public database created by Duke University that scientists, educators, students or the curious can mine for 3-D data on their species of interest.
"Each available CT scan provides an opportunity for the community to explore the morphology of a specimen. Those interested in the evolutionary diversity of vertebrates, as well as comparative and functional morphology, can quickly generate new research questions each time they look through high-resolution 3-D anatomical data," said U-M's Tucker, a professor of ecology and evolutionary biology and curator of mammals at the Museum of Zoology.
Advances in understanding the structure, function and evolution of genes and genomes have outpaced phenomics — the study of how genes interact with the environment to produce physical traits, or phenotypes. By providing a searchable digital encyclopedia of thousands of vertebrate phenotypes, oVert could be a valuable resource in narrowing the gap, Blackburn said.
"This is moving phenomes into Big Data science," he said.
CT scanning offers a wealth of data about specimens, but it's an expensive and time-intensive process, limiting the number of specimens that can be scanned, Blackburn said.
To maximize the usefulness of oVert, researchers will select and scan "super specimens," those that are representative of a species, are in good condition and have corresponding data in iDigBio, information such as where, when and by whom they were collected. Ideally, the specimen would also have genetic data.
One of the key features of oVert is making these data freely available, Blackburn said. In addition to easing researchers' access to vertebrate diversity, the oVert team also envisions the anatomical images being used in classrooms, inspiring the next generation of scientists, engineers and mathematicians.
The project, which kicks off Sept. 1, is not without its challenges: providing virtual access to specimens still requires boots-on-the-ground work. Specimens must be hand-selected, shipped, tracked, scanned, uploaded to MorphoSource and integrated with corresponding data in iDigBio. Multiply this process by 20,000 across 16 museum collections and the workflow can quickly become daunting, Blackburn said.
Sixteen research institutions are included in the grant: the University of Florida; U-M; Academy of Natural Sciences of Drexel University; California Academy of Sciences; Cornell University; Field Museum of Natural History; Harvard University; Louisiana State University; Scripps Institution of Oceanography at the University of California, San Diego; Texas A&M University; University of California, Berkeley; University of Kansas; University of Texas, Austin; University of Washington; Virginia Institute of Marine Science; and Yale University.
U-M was selected as a scanning center for the project because of the size and diversity of the Museum of Zoology's fluid-preserved vertebrate collections, as well as the availability of imaging equipment at the medical and dental schools. The priority will be to scan whole-body, fluid-preserved specimens containing articulated and complete skeletons, Tucker said.
The other U-M researchers involved in the project are Daniel Rabosky, Cody Thompson and Alison Davis Rabosky of the Museum of Zoology and the Department of Ecology and Evolutionary Biology. Rabosky is the lead principal investigator for U-M and Tucker is the co-principal investigator.
Duke University and the University of Chicago are also partners in the oVert project.