We discussed this nice preprint in journal club today, in which Taniel Winner (Berman, Ting, Kesar labs) comes up with a relatively simple procedure to identify and understand differences in walking gait between individuals (in this case, humans who had a stroke vs. those who did not). https://www.biorxiv.org/content/10.1101/2022.12.22.521665v1

They train recurrent neural networks to predict walking kinematics for each individual and then project the dynamical models onto a common low-dimensional representation of walking gait, which they can then use to compare "gait signatures" across individuals. This approach helps circumvent the issue that it is often tough to interpret differences in joint kinematics without an underlying model, but full biomechanical models are typically too intense.

Although it was only applied to humans in this preprint, I expect this approach will also be useful for those of us perturbing the nervous systems of flies/mice/monkeys/etc and trying to identify subtle and often confusing changes in body pose/joint kinematics.

My only criticism is that it needs a decent acronym. Best I came up with is U DISGUST ME (Unsupervised Discovery of Individual Stereotyped Gait Uniqueness Signatures To Manifold Everyone)

#Introduction
I'm an Assistant Professor at Emory University. Our lab uses the Drosophila taste system as a model to study sensory processing and modulation.

I'm passionate about teaching, outreach, and making science more inclusive and equitable. Outside of lab, you can find me trail running, baking, or consuming large amounts of baked goods.

Excited to connect here!

#introduction
Hi, #neuroscience community.

I'm an assistant professor at the University of Washington who works at the intersection of engineering and neuroscience. I am excited about using engineering tools like closed-loop motor brain-computer interfaces to probe basic neuroscience questions, and using insights about the brain to improve functionality of BCI applications.

My lab's current work
1) explores using BCIs as tools to begin unravelling how learning computations happen in neural populations
2) studies the interactions between users who learn alongside algorithms ("co-adaptation").
3) investigates how brains build "internal models", with a current interest in how to quantify internal model formation behaviorally.
We use both animal models (non-human primates) and human subject experiments. We also collaborate closely with computational and theoretical neuroscientists.

Trying out this thing as an alternative to Twitter. I'm already intimidated by the space to write and minimal pressure to edit.

#introduction
I am a postdoc at Stanford University, working with Krishna Shenoy. In collaboration with many experimental and computational colleagues, I study the neural mechanisms that control movement, and more broadly, how neural populations spanning interconnected brain regions perform the distributed computations that drive skilled behavior. I develop experimental and computational tools to understand the neural population dynamics that establish speed and dexterity.

I aim to discover insights into brain-wide computations in health and in neurological disease, with an eye towards identifying effective, targeted neuromodulation to treat movement disorders.

I also build open source tools:

• https://djoshea.github.io/neuropixel-utils/
• https://lfads.github.io/lfads-run-manager/
• https://github.com/djoshea/eraasr
• https://github.com/djoshea/haptic-control

Looking forward to joining the growing neuro community here!