Meet Thomas Baden, one of the PLOS Open Source Toolkit Channel Editors, and find out how he came to be a proponent of open source tools and a researcher working at the intersection of vision and neuroscience.
Can you tell me about yourself and your current research, and how you came to be interested in this field?
We are interested in how single and small groups of neurons compute information. To address aspects of this rather broad question, we work on early vertebrate vision, with focus on the retina where we use optical imaging, electrophysiology and computational modeling to interrogate genetically defined circuits. In particular, we are currently studying the relationship between neuronal computation and the natural input of the visual network. Different sighted vertebrates, such as mice, zebrafish or chicks, all used a similar fundamental blueprint of a retina network, consisting of the same classes of neurons. However, the complexity of their retinae differs dramatically, almost certainly a reflection of the different visual tasks that these animals faced during their evolutionary history. How then can you take an established, complex neuronal network like the retina and specifically “tune” it to efficiently deliver the type of visual information that matters to any one visual niche?
What made you decide to join the editorial team at the PLOS Open Source Toolkit Channel?
Apart from our work in vision science, we are also keen “tinkerers.” Much of our lab equipment is at least partially home built. There are 3D-printed parts all over the place, with blinking lights and cables sputtering about the place. Many of our solutions for our specific experimental needs are not unique, nor complicated. Others might benefit from these developments, as much as we already did benefit by adapting others’ solutions. We therefore believe that it is a key ingredient to scientific progress to share plans for all these tools and machines, small or big, for the greater benefit of everyone. After all, science is a fundamentally collaborative exercise, and that does not start with sharing scientific results. It starts by sharing the methods.
Where do you see your field of research heading in the next few years? What are the next big questions the field will address?
Over the past few years in particular, much progress has been made to map out the basic building blocks of early vision in a few model species. This work has led to a deep understanding of some aspects of neuronal computation. However, in many cases, we do not know how general these solutions are. If one species solves a particular computation in a particular way, does this mean that most other species do this, too? And how robust are these computations to the noisy and complex input that these systems need to deal with out in the wild? These types of questions are already beginning to intrigue researchers in the field, and likely we will see much advance in this area over the coming years.
What is your favorite piece on the Channel to date? What do you think is a must-read on the Channel?
I like the currently highlighted fluorescence chamber by Nunez et al. It’s so simple, but so effective. A handful of LEDs, some carefully chosen acrylic filter and a Raspberry Pi camera, and you get a high-performance, general purpose fluorescence chamber. The authors demonstrate how this is very useful for looking at bacterial cultures, but I strongly suspect that this would work for all kinds of stuff: Gel transillumination? Sorting transgenic flies/fish/worms with fluorescent markers? A very nice piece of work. One further nice point here is that the whole design does not, like so many, come from Western Europe or North America but from Chile, again highlighting the global nature of the open source hardware movement.
What is the importance of open source in your field?
I think it is very important and increasingly so. Let’s look at data alone. Many modern experimental techniques generate data at staggering rates now, much quicker than any one individual or lab can handle. Genome sequences, imaging data, or EM-resolution anatomy of entire brains—This type of data provides a fantastic opportunity for researchers anywhere to contribute to state-of-the-art science without having to invest in the typically extremely costly hardware required to generate the data.
For more open source tools check out the PLOS Open Source Toolkit