The Embryo Digital Atlas | #mozsprint 2017 Interview Series

Paul (@paulvilloutreix) is a Postdoctoral Fellow at Princeton University and a Visiting Fellow at the Center for Data Arts at The New School. I met Paul at our Working Open Workshop in Montreal where I was blown away at how he combines design, science and technology in his project, The Embryo Digital Atlas.

I interviewed Paul to learn more about the Embryo Digital Atlas and how you can contribute June 1–2 at #mozsprint.

What is The Embryo Digital Atlas?

The Embryo Digital Atlas is an open source web based platform to visualize complex experimental datasets of embryogenesis in an easy and beautiful way. Embryogenesis is the process by which a single fertilized egg is transformed into a multicellular organism. It is studied in various model organisms, from the sea urchin, to the fruit fly, to the zebrafish, to the chicken, to the mouse, to the human! Many mechanisms are conserved across species, this is why it is useful to study and compare this process in the various models. Imaging studies play a big part in these studies, mainly because this is such a complex and intricate process that researchers first need to be able to describe it comprehensively.

Embryogenesis is at the heart of the public debate about stem cell research or genome editing. In addition, it is a source of inspiration for engineers, designers and architects interested in how a complex structure can emerge from simple units. The aim of The Embryo Digital Atlas is to make public datasets of embryogenesis easily accessible for an audience ranging from curious citizen, to students, to professional researchers.

Why did you start The Embryo Digital Atlas?

While working with experimental biologists at Princeton University, I started noticing that there were many datasets of embryogenesis that were sitting in the labs and were not necessarily published because they were not associated to positive results. I thought that they could be useful as public resources for the general understanding of embryogenesis.

At the same time, I was playing with some image datasets and found that the current visualization tools were either too generic or required a lot of expert knowledge. To find ways of making scientific image datasets easily and interactively accessible, I started to collaborate with designers and artists at the Center for Data Arts at The New School, leading to The Embryo Digital Atlas project.

Why is it important to be able to visualize embryogenesis?

Embryogenesis is a very complicated process. It involves an enormous amount of variables that need to monitored at the same time to just begin to understand what the correlations between all the processes are. It involves 3D changes in shape that can be challenging to interpret if you cannot interact with the data. Microscopes keep improving, bringing more information, resolution and larger datasets. The study of embryogenesis currently revolves around visualizations. We are at this stage in the field where most of the scientific evidence is based on images.

What problems have you run into while working on this project?

As project lead, the main difficulty is designing specific short term problems and issues that can be solved while maintaining a working prototype and having a project that is consistent with long term goals.

We started developing the platform in JavaScript using the library three.js which implements WebGL. This library is really user friendly and we could quickly have a support for drawing images that can be moved in a 3D space while handling most of the user interactions. The current datasets are relatively small, about 30–40 MB. One problem that we are currently facing is how to visualize much larger datasets, from 10 GB to a few TB.

Later on we would like to combine this visualization tool to an image storage platform. The Open Microscopy Environment is a great community developing open source software which should be very useful for this part of the project. Other future developments involve collaborative annotation of the datasets or using Virtual Reality to explore immersive approaches to scientific data visualization.

What kind of skills do I need to help you?

My main goal at the Mozilla Global Sprint is to set up an operational website, with a nice overall interface,smooth navigation,content that is well organized among the pages and some nice features such as animations during loading times. I am looking for anyone with skills in user interface and web design who would like to contribute to a cool science project!

How can others join your project at #mozsprint 2017?

A few easy things that could be very helpful to start with are contributions to the layout, particularly the implementation of the header, footer, tabs and navigation. It could also be great if the layout would be adaptable to tablets and smartphones with smaller screens. That requires some knowledge in html, css, and php and some design skills.

The project is currently on my personal github — https://github.com/paulvill/paulvill.github.io. There is a list of issues which is maintained up-to-date on the github repository, so do not hesitate to jump on any where you think you can contribute.

I can be reached on twitter at @paulvilloutreix or by email at paul.villoutreix@gmail.com


Join us wherever you are June 1–2 at the Mozilla Global Sprint to work on The Embryo Digital Atlas and many other projects! Join a diverse network of scientists, educators, artists, engineers and others in person and online to hack and build projects for a health Internet. Get your tickets now!

This is cross-posted on the Mozilla Open Leaders blog.