Austin E. Y. T. Lefebvre
@austin-lefebvre.bsky.social
Dr. Leafbeaver, DS at Calico.
You can't spell imaging without "im aging" and boy am I.
Organelle enthusiast.
Creator of Nellie and Mitometer.
You can't spell imaging without "im aging" and boy am I.
Organelle enthusiast.
Creator of Nellie and Mitometer.
Reposted by Austin E. Y. T. Lefebvre
Analyzing organelles is HARD. They're complex, but critical to understanding biology. Current tools are manual, organelle-specific, or can't handle large 3D datasets.
With an easy install and a click of a button, Nellie solves ALL these problems!
nature.com/articles/s41...
🧵2/N
With an easy install and a click of a button, Nellie solves ALL these problems!
nature.com/articles/s41...
🧵2/N
Nellie: automated organelle segmentation, tracking and hierarchical feature extraction in 2D/3D live-cell microscopy - Nature Methods
Nellie is a comprehensive automated pipeline for studying the structure and intracellular dynamics of diverse organelles that offers accurate segmentation, tracking and feature extraction on both 2D a...
nature.com
February 27, 2025 at 8:13 PM
Analyzing organelles is HARD. They're complex, but critical to understanding biology. Current tools are manual, organelle-specific, or can't handle large 3D datasets.
With an easy install and a click of a button, Nellie solves ALL these problems!
nature.com/articles/s41...
🧵2/N
With an easy install and a click of a button, Nellie solves ALL these problems!
nature.com/articles/s41...
🧵2/N
This is definitely my favorite response yet haha, my favorite movie of the year!
March 10, 2025 at 9:25 PM
This is definitely my favorite response yet haha, my favorite movie of the year!
Thank you!!
March 10, 2025 at 8:58 PM
Thank you!!
Thanks a lot Nico!
March 10, 2025 at 8:58 PM
Thanks a lot Nico!
Thanks again Felipe!
March 10, 2025 at 8:43 PM
Thanks again Felipe!
Thanks for the post! Hope everyone finds it useful. And don’t hesitate to reach out if you have any questions or issues!
March 6, 2025 at 5:28 PM
Thanks for the post! Hope everyone finds it useful. And don’t hesitate to reach out if you have any questions or issues!
Great collection. Thanks for including Nellie!!
March 3, 2025 at 6:46 AM
Great collection. Thanks for including Nellie!!
Thanks a lot Darren!
February 27, 2025 at 9:33 PM
Thanks a lot Darren!
Try and share Nellie today!
We've included some sample data if you want to explore what Nellie can do.
Code: github.com/aelefebv/nel...
Paper: nature.com/articles/s41...
And please share your results with us! We're excited to see how you'll use it in your research.
🧵Fin/N
We've included some sample data if you want to explore what Nellie can do.
Code: github.com/aelefebv/nel...
Paper: nature.com/articles/s41...
And please share your results with us! We're excited to see how you'll use it in your research.
🧵Fin/N
GitHub - aelefebv/nellie: Nellie: Automated organelle segmentation, tracking, and hierarchical feature extraction in 2D/3D live-cell microscopy
Nellie: Automated organelle segmentation, tracking, and hierarchical feature extraction in 2D/3D live-cell microscopy - aelefebv/nellie
github.com
February 27, 2025 at 8:13 PM
Try and share Nellie today!
We've included some sample data if you want to explore what Nellie can do.
Code: github.com/aelefebv/nel...
Paper: nature.com/articles/s41...
And please share your results with us! We're excited to see how you'll use it in your research.
🧵Fin/N
We've included some sample data if you want to explore what Nellie can do.
Code: github.com/aelefebv/nel...
Paper: nature.com/articles/s41...
And please share your results with us! We're excited to see how you'll use it in your research.
🧵Fin/N
We also compare Nellie to existing state of the art methods for both segmentation and tracking across imaging resolution, structure widths, and structure lengths, as well as the pipeline's speed on different sized images
🧵14/N
🧵14/N
February 27, 2025 at 8:13 PM
We also compare Nellie to existing state of the art methods for both segmentation and tracking across imaging resolution, structure widths, and structure lengths, as well as the pipeline's speed on different sized images
🧵14/N
🧵14/N
Now we can generate flow fields and get cool metrics, like if fields are diverging from (fission?) or converging to (fusion?) a point, or whether a branch is rotating along its long axis (?? dancing?). All of this, and much much more, linked to our organellar hierarchy.
🧵13/N
🧵13/N
February 27, 2025 at 8:13 PM
Now we can generate flow fields and get cool metrics, like if fields are diverging from (fission?) or converging to (fusion?) a point, or whether a branch is rotating along its long axis (?? dancing?). All of this, and much much more, linked to our organellar hierarchy.
🧵13/N
🧵13/N
From these beacons, we can use a distance and cost-weighted interpolation to generate tracks for any coordinate within our organelles! Even at a sub-voxel level. We can use this to link voxels (and thus objects) between temporal frames too.
🧵12/N
🧵12/N
February 27, 2025 at 8:13 PM
From these beacons, we can use a distance and cost-weighted interpolation to generate tracks for any coordinate within our organelles! Even at a sub-voxel level. We can use this to link voxels (and thus objects) between temporal frames too.
🧵12/N
🧵12/N
We generate internal motion capture markers (think Hollywood) and pattern match these markers between frames to create beacons that point the way for the actual tracking. This means that we don't have to rely on finicky segmentations or skeletons for tracking.
🧵11/N
🧵11/N
February 27, 2025 at 8:13 PM
We generate internal motion capture markers (think Hollywood) and pattern match these markers between frames to create beacons that point the way for the actual tracking. This means that we don't have to rely on finicky segmentations or skeletons for tracking.
🧵11/N
🧵11/N
Then Nellie breaks down the organelles into connected components, and uses its skeleton to break it down even further into branches, which break down into individual nodes, which are variable-range containers for surrounding voxels. A big hierarchical organelle tree.
🧵10/N
🧵10/N
February 27, 2025 at 8:13 PM
Then Nellie breaks down the organelles into connected components, and uses its skeleton to break it down even further into branches, which break down into individual nodes, which are variable-range containers for surrounding voxels. A big hierarchical organelle tree.
🧵10/N
🧵10/N
So how does Nellie actually work? First Nellie figures out your image's metadata. Then it applies an automated structural-enhancement regime (modified Frangi filter + others) to make those organelles pop.
🧵9/N
🧵9/N
February 27, 2025 at 8:13 PM
So how does Nellie actually work? First Nellie figures out your image's metadata. Then it applies an automated structural-enhancement regime (modified Frangi filter + others) to make those organelles pop.
🧵9/N
🧵9/N
Before deep-diving into Nellie, shout out my co-authors: @gavsturm.bsky.social , for help on all the paper, experiments, and testing, and to the other co-authors @kayleyhake.bsky.social, Ben, Emily, Magdalena, and Molly.
And of course @ritastrack.bsky.social for being a great editor!
🧵8/N
And of course @ritastrack.bsky.social for being a great editor!
🧵8/N
February 27, 2025 at 8:13 PM
Before deep-diving into Nellie, shout out my co-authors: @gavsturm.bsky.social , for help on all the paper, experiments, and testing, and to the other co-authors @kayleyhake.bsky.social, Ben, Emily, Magdalena, and Molly.
And of course @ritastrack.bsky.social for being a great editor!
🧵8/N
And of course @ritastrack.bsky.social for being a great editor!
🧵8/N
And just for fun, here's me walking around, masked by Meta's Segment Anything Model and tracked via Nellie's tracking pipeline
🧵7/N
🧵7/N
February 27, 2025 at 8:13 PM
And just for fun, here's me walking around, masked by Meta's Segment Anything Model and tracked via Nellie's tracking pipeline
🧵7/N
🧵7/N
We used Nellie to unmix multiple organelles from a single fluorescence channel with high accuracy using just their morphology and motility patterns with a standard (random forest) classifier.
More fluorescence channels for your experiments!
🧵6/N
More fluorescence channels for your experiments!
🧵6/N
February 27, 2025 at 8:13 PM
We used Nellie to unmix multiple organelles from a single fluorescence channel with high accuracy using just their morphology and motility patterns with a standard (random forest) classifier.
More fluorescence channels for your experiments!
🧵6/N
More fluorescence channels for your experiments!
🧵6/N