The Card lab at Columbia Univ (NYC) is hiring a mechanical engineer! If you like tinkering with mechanical systems and want to help us build custom rigs for neuroscience research, this job is for you! Please feel free to pass along to interested folks.
hhmi.wd1.myworkdayjobs.com/en-US/Extern...
Behavioral biases drift in individuals.
β’Β Drift happens on many timescales
β’ Drift rate depends on genetics and neuromodulators
β’ It might be adaptive for rapidly changing environments
elifesciences.org/reviewed-pre...
π Thrilled to see our #MosquitoCellAtlas featured on the cover of @cp-cell.bsky.social ! This mosaic represents 367K cells across 19 tissues, unveiling the complex biology that makes mosquitoes the world's deadliest animals. Beautiful art by @somedonkey.bsky.social meets the #Vosshalllab science! π¦π¨
It might actually work better on humans as the original training dataset contains a lot more humans than ants videos!
Thank you! We have not tried on humans ourselves but there is no reason why FERAL could not work on human behavior videos! We just like ants videos more (who wouldn't?)
That sounds cool! Feel free to reach out if you have any questions! We would love to see the videos of hermit crabs
We dooooo
Finally me and Peter would like to thank all our collaborators and colleagues who helped! Especially @leslievosshall.bsky.social and the members of the Vosshall lab! This work would have never been possible without their support, help and passion for animals (with a obvious bias toward mosquitoes)
But what we are really interested in seeing is what sort of behaviors and recordings the community will try it on!
Check out FERAL
Preprint: www.biorxiv.org/content/10.1...
Code + Docs: www.getferal.ai
Feel free to contact us with any feedback, question or suggestion!
π Emergent collective behavior
Finally, with @tomaskay.bsky.social we asked whether FERAL can detect collective dynamics directly from raw frames.
We annotated raiding phases in colonies of ants.
FERAL identifies the onset and duration of colony raids and collective behavior without tracking!
π¦ Camera-trap footage of chimpanzees & gorillas (PanAf500)
We pushed FERAL to handle multi-species, multi-behavior field recordings: day and night, dense vegetation, variable lighting.
FERAL achieves strong accuracy even for visually rare but distinct behaviors like climbing up/down.
This is also a great example of how tracking and FERAL are complementary:
β‘οΈ track the animal to get WHERE it is
β‘οΈ apply FERAL to its video crop to get WHAT itβs doing
(We think this two-step approach has great potential, so if you are interested in this stay tuned)
FERAL goes wild:
π¦ Drone footage of Grevyβs zebras (with @blaircostelloe.bsky.social, Ben Koger and @icouzin.bsky.social)
Pose-estimation pipelines struggled to detect vigilance behavior from overhead drone videos.
FERAL identifies vigilance bouts directly from video.
Ant social interactions (with Janet Zhao & Dominic Frank)
Next we wanted to try FERAL on recordings of social behavior. FERAL detects both self-grooming and allogrooming directly from raw video, even with occlusions and two animals with very different shapes.
How does FERAL perform on lab recordings?
πͺ± C. elegans locomotion (with Friederike Buck)
FERAL incorporates temporal dynamics in its learning. In this dataset, it separates forward and reverse crawling: behaviors that look nearly identical frame-to-frame, but are revealed by temporal dynamics!
Performance
On benchmarks, FERAL beats both pose- and video-based baselines: 94.5% mAP on CalMS21 (mouse social interactions). Outperforms Googleβs VideoPrism while using only 25% of the data.
Hereβs a snippet of FERAL segmenting mouse social behavior:
Give FERAL:
1. videos
2. behavioral annotations
FERAL fine-tunes a video foundation model on your data.
FERAL outputs:
- segmented ethograms
- model weights you can reuse
All in few lines of code (no ML or coding expertise needed). It runs easily on Google Colab.
Try it yourself: www.getferal.ai
Our preprint is out! www.biorxiv.org/content/10.1...
Peter Skovorodnikov and I are excited to present FERAL: a new video-understanding toolkit that maps raw video directly to behavior, no pose estimation required.
It works across species, from lab to field, and even in collective systems. (π§΅1/n)
#sfn2025 three fun posters from us today: theory work by @rgast.bsky.social on the effects of neural heterogeneity on computation, a hippocampus+feeding collab with @liye-tsri.bsky.social , and LLMs for pose interpretation in rehabilitation data with @peabody124.bsky.social
#SfN25 stop by board ZZ5! We are presenting FERAL!
As you can see in this pic we practiced this presentation a lot and we also have some FERAL tshirts! #getferalatSfN
#SfN25 Monday morning we present FERAL: a video-understanding tool for animal behavior detection without the need for tracking or pose estimation! FERAL detects single animal, social and collective behavior in the lab and the wild! Visit our poster at board ZZ5 and check more here: www.getferal.ai
Excited to announce the final version of the Mosquito Cell Atlas is out now in @cellpress.bsky.social!! π¦π©Έ
There is SO much left to find & investigate in this dataset (& the rich biology of the Aedes aegypti mosquito)! We hope this helps scientists in many fields!
www.cell.com/cell/fulltex...
The male genitalia of an ππ¦π₯π¦π΄ π’ππ£π°π±πͺπ€π΅πΆπ΄ mosquito
What does mating look like when you only have a single shot at getting it right?
Very excited to share our work on an almost-invisible female control, rapidly evolving mating recognition systems, and species that break the rules and take over the world. IN MOSQUITOES>
New
@currentbiology.bsky.social paper from @leslievosshall.bsky.social&@leahhouri.bsky.social: FEMALE mosquitoes control mating, not males, by special genital movement. No elongation = no mating! Plus: male asian tiger mosquitoes can bypass the yellow fever mosquito female control shorturl.at/tsIWH
This is extended data figure 9, which shows a reconstruction of the plumage of Archaeopteryx by Michael Rothman.
A paper in Nature presents a suite of previously unknown features belonging to the oldest known fossil bird, Archaeopteryx. Analysis of the Chicago Archaeopteryx shows new information about the skeleton, soft tissues and plumage of this iconic taxon. go.nature.com/4jNZIJg #Paleosky π§ͺ
#πͺΆ
The Rockefeller Uni ducklings are back!
#birds
#birds
