Improving Transparency with protocols.io
Author: Wyatt Million
Wyatt is a PhD candidate studying marine biology at the University of Southern California. His research focuses on the role that phenotypic plasticity plays in coral ecology and evolution. His career in science has been driven by his desire to conserve and restore natural environments as well as his love for the outdoors. He uses a combination of field and lab-based experimentation to identify patterns in the responses of endangered coral to changing environments in an effort to understand and predict species persistence in the future.
From my own experience and the experience of other PhD students, I have realized that we rely heavily on previously published work to learn new methods. Publications that provide the data and directions to recreate the analysis are a gold mine to a student looking to apply that method or analysis to their own research. However, there have been numerous times where the perfect paper comes out, but when I try out the analysis it is nearly impossible to recreate it from the limited description in a methods section. There have been efforts in computational biology to improve access by making code and data publicly available, however similar attempts are less common for wet-lab procedures. To support this open-access movement, my advisor and I felt a strong need to make our field and wet-lab protocols as transparent and accessible as possible, which is why I recently published them on protocols.io. We tried to include everything a reader might need for learning 3D coral photogrammetry, so they can begin measuring coral with their own 3D models.
A large component of my PhD research is focused on a field-based experiment where we track the growth of hundreds of endangered coral colonies transplanted to reefs in the Florida Keys Reef Tract. Initially, coral colonies were small and easy to measure by hand but as expected, they have grown into complex 3D structures where traditional methods are arduous if not impossible. In an effort to overcome this hurdle, I developed a method to create, then measure 3D models of these corals.
When I started developing my method, 3D photogrammetry was common in other fields and was emerging in coral biology. While publications existed that used this technology, no clear protocols were available and I ended up developing my own method through sparse guidance and a lot of trial and error.
I refined my method during critical data collection trips to our field sites. This game-time testing came at a cost of usable data, especially for early time-points, but it was really the only way to work on the method due to limited time in the field. I realized that having a tried and true protocol prior to these early field trips would have improved my data collection drastically, however there was no protocol to guide me. Instead, I incrementally improved my method and, thanks to a patient advisor, was able to create a final protocol that I knew would give me results. Now, I am able to train divers in 3D photogrammetry in a single day so that they can collect photographs to produce excellent 3D models. I wanted to share what I had created so that others wouldn’t have to reinvent the wheel to be able to use 3D photogrammetry in their own research. This is why I published my protocols on protocols.io.
protocols.io gave me a platform to share detailed steps for 3D photogrammetry in a way that a journal publication wouldn’t allow. There are many intricacies that are left out of formal methods sections that are potentially vital to fellow researchers hoping to learn, recreate, and implement a specific technique. I hoped to provide all of the information needed for other coral biologists to use 3D photogrammetry in their own work without having to struggle as I did. Beyond a stand-alone protocol, I can now link my published protocols to any future manuscripts so that readers can have access to a set of directions and examples to work through the methods I outline in the article.
As a graduate student not just reading, but learning from those who came before me, I urge researchers to be as transparent as possible in their methods. One way to do that is through protocols.io. I published our protocols because I truly wanted to make 3D photogrammetry accessible in coral biology. For those who are using a new or improved protocol, I suggest doing the same!
Referenced Links:
Phenotyping 3D coral models in MeshLab
https://www.protocols.io/view/phenotyping-3d-coral-models-in-meshlab-bgbpjsmn
Image capture and pre-filtering for 3D photogrammetry of coral colonies
https://www.protocols.io/view/image-capture-and-pre-filtering-for-3d-photogramme-bgdcjs2w
