Datta Lab

…on the development and use of Structured Variational Autoencoders to better understand behavioral (and other classses) of data, posted today to the arXiv. The paper is super-cool – it describes a method to learn representations of complex time-series data (which are often non-linear) that are well-suited to structured models (using a Hidden Markov Model, for example). The approach could be used to solve many practical problems in modeling behavioral data, and we are really excited to try it on our 3D datasets. Congrats to Matt and Alex (and of course Ryan Adams) on their great work!

…can be found here.

We made a little movie explainer for the public about our new approach to behavioral characterization. We hope you like it!

…then check out the latest paper from the lab in this week’s issue of Neuron! The paper describes a new technology for behavioral characterization and classification that combines 3D machine vision with recently developed approaches in unsupervised machine learning. This combined method can be used to objectively characterize mouse behavior in a wide variety of circumstances, and to identify changes in action induced after manipulations of the environment, the genome or neural activity; as such, this approach promises to provide fundamental insight into the complex relationships between the genome, neural activity and patterns of action. We are currently using this method in the lab in a wide array of experiments, ranging from characterizing naturalistic patterns of odor-driven behavior in rodents to trying to identify nodes within corticostriatal circuits responsible for behavioral sequencing. Congrats to Alex, Matt and others in the lab who contributed to the paper!

(PDF can be found under the “Publications” tab, and further discussion in the “Research” section).

Thank god that is over! 30,000 people, and a seemingly equal number of posters. Also answered was the question: how many slices of deep dish can one eat in a week? Stan, Alex and Ralphie did a great job sharing their data with the community, our colleagues Bernardo and Beth gave fantastic plenary talks, and it was terrific to see Mike and Carla win the Gruber prize. Enough to make this an annual event? Maybe…..

Alex, somewhere in the scrum, talking about our new 3D behavioral analysis technology

Blurry but well-earned: Carla and Mike accepting the Gruber prize!

I appreciate the frankness of the vomit fee in Chicago cabs

Ralphie’s first real meeting poster

A hearty N’ Sync farewell to our intrepid now former lab manager, Ally Nowlan! Ally was an amazing member of the lab and our backbone for several years – we wish her that absolute best of luck as she starts grad school with the Cold Spring Harbor crew in neurobiology. Steve Shea has already been warned!

To celebrate her contributions to the lab and her new career as a graduate student, we did a little karaoke. We learned that, no matter where in the world you come from, you almost certainly enjoy drunk singing to Bohemian Rhapsody. We also welcomed our new lab manager, Neha Bhagat, who until recently was an undergraduate at Northeastern and had spent time in the Datta lab diligently injecting the brains of animals with viruses, amongst other things. Ally you will be missed – keep in touch, and good luck with everything! And a hearty welcome to Neha!

Ally wishing us all Bye Bye Bye

Neha, with pipette at the ready!

…come to this great smell meeting organized by Anne Marie Oswald and Bob at UCSB in collaboration with the Kavli Institute of Theoretical Physics this summer. Should be a blast, and at least part of the Datta lab will represent. Going to be a blast!

The Datta Lab (plus a couple F1s), at Good Harbor Beach.

Waiting for the lobster rolls. You can’t really tell, but it is COLD!

New Postdoc Tatsuya. Not on TV (yet).

Summer Student Max. Currently on TV.

A happy fall (and end of summer) from the Datta lab! Lots has happened in the last 18 months (the approximate timescale upon which this blog currently operates), nearly none of which was documented, so this is an attempt to compress some of the highlights into a single blog entry. MUCH will be left unsaid – our apologies to those affected by any oversights.

First, we went to the beach (see above)! This was awesome, since the rip tides were in full effect and the waves were like something out of the movie Sharknado. We didn’t lose anyone, and those that stuck it out ended up playing mini-golf on Route 1 in Saugus in front of a plastic green dinosaur. Totally worth it.

Second, the awards roll – postdoc Giuliano Iurilli won a prestigious Human Frontiers fellowship to work in the lab on the striatum, graduate student Tari Tan netted an F31 NRSA for her work on non-canonical olfactory systems, new postdoc Tatsuya Tsukahara joined us from the University of Tokyo (where he did spectacular work in developmental molecular biology in Hiroyuki Takeda’s and Yoshi Watanabe’s labs) armed with a Japan Society for the Promotion of Science fellowship, and Bob was recently named a Vallee Foundation Young Investigator.

Third, postdoc Andrew Giessel wrote a really nice review on odor coding and circuit function different olfactory regions in the brain, linked here. Check it out!

Finally, the lab also got bigger, despite the fact that Dan recently decided that aerobic exercise is a “thing” and Bob has more or less stayed low carb. Joining our awesome Northeastern Co-op student Ralph Peterson (who has been working with Alex on behavior) are three more Co-op students – Jesse Katon (also working on new methods for behavioral analysis), Neha Bhagat and Alex Williams (both working to deliver optogenetic reagents to across the olfactory cortical mantle). This summer saw us visited by Max Burkholder, who plays Max Braverman on the TV show Parenthood, and who worked with Ally doing molecular biology. All these new people needed some place to work so we (temporarily) built a lab expansion next to our new colleague Matt Pecot’s lab to get some elbow room.

OK, enough blogging for now. Back to science!