Find research datasets worth reusing
Search datasets from major research repositories and use ShareScore to quickly assess how well each record supports discovery, access, and reuse.
13
datasets available to search
ShareScore release 0.9.0
Dataset results
13 results for “deeplabcut”
DeepLabCut: markerless pose estimation of user-defined body parts with deep learning
<p>This data entry contains <strong>annotated mouse data from the <a href="https://www.nature.com/articles/s41593-018-0209-y">DeepLabCut Nature Neuroscience paper</a></strong>.</p> <p>This data entry contains a public release of annotated mouse data from the DeepLabCut paper. The trail-tracking behavior is part of an investigation into odor guided navigation, where one or multiple wildtype (C57BL/6J) mice are running on a paper spool and following odor trails. These experiments were carried out by Alexander Mathis & Mackenzie Mathis in the Murthy lab at Harvard University. </p> <p>Data was recorded by two different cameras (640×480 pixels with Point Grey Firefly (FMVU-03MTM-CS), and at approximately 1,700×1,200 pixels with Grasshopper 3 4.1MP Mono USB3 Vision (CMOSIS CMV4000-3E12)) at 30 Hz. The latter images were cropped around mice to generate images that are approximately 800×800. </p> <p>Here we share 1066, frames from multiple experimental sessions observing 7 different mice. Pranav Mamidanna labeled the snout, the tip of the left and right ear as well as the base of the tail in the example images. The data is organized in <a href="https://www.nature.com/articles/s41596-019-0176-0">DeepLabCut 2.0 project structure</a> with images and annotations in the labeled-data folder. The names are pseudocodes indicating mouse id and session id, e.g. m4s1 = mouse 4 session 1.</p> <p>Code for loading, visualizing & training deep neural networks available at <a href="http://https://github.com/DeepLabCut/DeepLabCut"> https://github.com/DeepLabCut/DeepLabCut</a>.</p>
DeepLabCut network trained to track mouse 'front' body parts during rotarod running (front-view)
<p>DeepLabCut (https://github.com/DeepLabCut/) (Mathis et al., 2018; Nath et al., 2019) was used for tracking body parts of mice in an open field arena or in the rotarod. DeepLabCut 2.1.8.2 (local version on Windows with CPU, using the GUI) and 2.1.10.2 (google colab to train the network) were used using default parameters and the pretrained resnet50 network with imgaug augmentation. Frames were extracted with the k-means method and outlier frames with the jump method. <em>Rotarod, front camera: </em>29 frames from 18 videos (10 fps) were extracted for a total of 520 labeled pictures. 12 body parts (left, right and mid snout, left/right top/bottom ears, left/right eyes, headmount, left/right foot), 4 corners of the rotarod and 4 points on the rotarod wheels were manually labeled and linked to each other using skeletons. A neural network was trained using these images for 225K iterations (train error: 1.58, test error: 1.63). 20 outlier frames were extracted from each video and relabeled. An additional 20 images from 20 new videos with different recording conditions were labeled. The network was then refined for 331K iterations (from scratch) (train error: 2.34, test error: 5.49). This process was repeated a second time when adding 20 new videos (400 frames) and the network trained to a final 402K (train error: 2.64, test error: 3.98). For this last batch, brightness/contrast were too low to detect body features; brightness/contrast were thus enhanced using custom-written Python scripts. Relevant videos were analyzed at each of the 3 steps, for a total of 152 videos.</p> <p><em>Used to analyze videos for a publication (Labouesse et al., Nature Communications 2023).</em></p> <p><em>Network not included in the final publication.</em></p>
DeepLabCut network trained to track mouse body parts during open field locomotion (top-down view)
<p>DeepLabCut (https://github.com/DeepLabCut/) (Mathis et al., 2018; Nath et al., 2019) was used for tracking body parts of mice in an open field arena or in the rotarod. DeepLabCut 2.1.8.2 (local version on Windows with CPU, using the GUI) and 2.1.10.2 (google colab to train the network) were used using default parameters and the pretrained resnet50 network with imgaug augmentation. Frames were extracted with the k-means method and outlier frames with the jump method. <em>Open field: </em>20 images from 19 videos (10 or 30 fps) were extracted for a total of 380 labeled pictures. 8 body parts (snout, both ears, body center, both side laterals, tail base and tail end) and the 4 corners of the field arena were manually labeled and linked to each other using skeletons. A neural network was trained using these images for 170K iterations. 20 outlier frames were extracted from each video and relabeled. An additional 20 images from 19 videos with different recording conditions were labeled. The network was then refined for 210K iterations (from scratch), yielding a train error of 3.33 pixels and a test error of 8.83 pixels (with a likelihood p-cutoff of 0.6). This process was repeated a second time (using an additional 20 images from 15 new videos) to improve the pixel error; to a final 400 K iterations (train error: 2.65, test error: 3.71). 67 videos from 5 different experiments were analyzed on the final network.<em> </em></p> <p><em>Used to analyze videos for a publication (Labouesse et al., Nature Communications 2023)</em></p>
DeepLabCut network trained to track mouse 'back' body parts during rotarod running (back-view)
<p>DeepLabCut (https://github.com/DeepLabCut/) (Mathis et al., 2018; Nath et al., 2019) was used for tracking body parts of mice in an open field arena or in the rotarod. DeepLabCut 2.1.8.2 (local version on Windows with CPU, using the GUI) and 2.1.10.2 (google colab to train the network) were used using default parameters and the pretrained resnet50 network with imgaug augmentation. Frames were extracted with the k-means method and outlier frames with the jump method. <em>Rotarod, back camera:</em> 20 images from 9 videos (10 fps) were extracted for a total of 180 labeled pictures. 5 body parts (2 paws, 2 ankles, tail base), 4 corners of the rotarod, 2 points on the rotarod wheels and 4 points in a flashing LED (indicating timestamps) were manually labeled. A neural network was trained using these images for 80K iterations. 20 images from 14 videos with different recording conditions were labeled. The network was trained to 200K iterations (from scratch) (train error: 3.00, test error: 3.75). Relevant videos were analyzed at each step, for a total of 152 videos.</p> <p><em>Used to analyze videos for a publication (Labouesse et al., Nature Communications 2023)</em></p>
Dopamine activity in the tail of the striatum, DeepLabCut and MoSeq during novel object exploration
<p>In this study, we characterized dynamics of novelty exploration using multi-point tracking (DeepLabCut) and behavioral segmentation (MoSeq). Mice were habituated in an arena, and then a object was placed at the corner of the arena. We compared 4 groups of mice: one with presentation of a novel object (stimulus novelty), one with a presentation of a familiar object (contextual novelty), one with presentation of a novel object after ablation of dopamine neuorns that project to the tail of the striatum (TS), and one with presentation of a novel object after sham surgery. With a separate group of mice, dopamine activity in TS was recorded during novelty exploration.</p>
Odor trail tracking data from DeepLabCut preprocessed for I-MuPPET
<p>This data entry contains <strong>annotated mouse data </strong>from the<strong> <a href="https://www.nature.com/articles/s41593-018-0209-y">DeepLabCut Nature Neuroscience paper</a></strong> preprocessed for <a href="https://urs-waldmann.github.io/i-muppet/"><strong>I-MuPPET</strong></a>.</p> <p>Code for I-MuPPET available at <a href="https://github.com/urs-waldmann/i-muppet/">https://github.com/urs-waldmann/i-muppet/</a>.</p>
Dopamine activity in the tail of the striatum, DeepLabCut and MoSeq during novel object exploration
Open the record for dataset details and reuse information.
KineWheel-DeepLabCut
<p>Video recordings:</p> <ol> <li><strong>KineWheel-DeepLabCut Setup Recording </strong>(2022-08-19T14-07-11--Basler acA720-520uc-UV-x264.mp4)<br> This video provides a recording produced by a KineWheel-DeepLabCut setup. It demonstrates a mouse running on a wheel, with frames alternately illuminated by UV and white light.</li> <li><strong>White-Light Illuminated Frames</strong> (2022-08-19T14-07-11--Basler acA720-520uc-x264.mp4)<br> This video is a clip extracted from the original recording, containing only frames that were illuminated by white light. These frames were subsequently used as input to the neural network.</li> <li><strong>Neural Network Output with Overlayed Annotations</strong> (2022-08-19T14-07-11--Basler acA720-520uc-x264-labeled.mp4)<br> This video presents the output of the neural network. It shows the same video clip as in Video 2, but with annotations predicted by the network overlaid on the frames.</li> </ol> <p>Along with the videos, we are also providing two sets of annotation coordinates: </p> <p><strong>CSV File: Neural Network Annotation Coordinates</strong><br> This CSV file contains the coordinates of the annotations predicted by the neural network and overlaid on the frames in Video 3.</p> <p><strong>HDF5 File: Neural Network Annotation Coordinates</strong><br> This HDF5 file also contains the coordinates of the annotations predicted by the neural network. These annotations correspond to the overlays presented in Video 3.</p>
DeepLabCut network for pupillometry analysis (Privitera et al. 2020)
<p>Network trained for Privitera et al. 2020. Used for the analysis of pupillometry video files using an infrared camera and anesthetized mice</p>
Videos of mice and squirrels handling various seeds and trained DeepLabCut models for tracking the digits
<p>Videos of freely moving and head-fixed mice handling flaxseeds, wheat berries, millet grains, couscous pellets, black-eyed peas, and peanut fragments, as well as wild squirrels handling peanuts, as used in Barrett et. al., "Manual dexterity of mice during food-handling involves the thumb and a set of fast basic movements", in preparation, plus trained DeepLabCut (DLC) models for tracking the digits of the mice during behaviour.</p> <p>The ZIP file "seed handling videos.zip" contains the videos analysed in the paper as H.264-encoded MP4s with a constant rate factor of 20, which has been shown to not significantly affect DLC tracking results (https://www.biorxiv.org/content/10.1101/457242v1). The ZIP file "squirrel seed handling videos.zip" contains the squirrel peanut handling videos in the same format. Each remaining ZIP file contains one or more DLC project folders (minus the videos subfolder), each containing a trained model for tracking the nose and digits 1-4 on both hands during seed handling. SeedHandling-John-2019-07-02.zip is for front view videos, SeedHandling-John-2019-07-04.zip bottom view, and SeedHandling-John-2019-07-05.zip head-fixed side view. "squirrel deeplabcut.zip" contains models for each set of squirrel videos (SeedHandling-John-2019-03-15\ for 20180921*.mp4 and SeedHandling-John-2019-05-20\ for20190424*.mp4). To use on your own data, simply unzip and use the analyze_videos function in DeepLabCut, setting the config file path to the folder you unzipped the model to. For more details on DeepLabCut, see Mathis et. al. (2018), Nat. Neurosci (https://www.nature.com/articles/s41593-018-0209-y) and http://www.mousemotorlab.org/deeplabcut.</p>
Videos for deeplabcut, noldus ethovision X14 and TSE multi conditioning systems comparisons
<p>Supplementary top view videos of OFT (open field test), EPM (elevate plus maze) and FST (foreced swim test) of adult male mice for deeplabcut, noldus ethovision X14 and TSE multi conditioning systems comparisons. All code used for analysis of these videos can be found under: https://github.com/ETHZ-INS/DLCAnalyzer. Additionally, DeepLabCut output files (point tracking data) of the videos and manual annotation of behaviors can be found in the same github repository. Additionally, the publication associated with this data here can be found under https://www.biorxiv.org/content/10.1101/2020.01.21.913624v1</p>
neuralnetwork_deeplabcut
Drawing uploaded to scidraw.io on: 08 May 2020
neuralnetwork_deeplabcut
Drawing uploaded to scidraw.io on: 08 May 2020
ScienceDex guides
Understand access before you commit
These curated guides explain access requirements, typical timelines, costs, and reuse considerations for widely used research datasets.
Allen Brain Atlas
Allen Brain Atlas is an Allen Institute collection of brain map atlases, datasets, APIs, and analysis tools covering mouse, human, and non-human primate brain resources.
Annotated Behaviour and Observability Dataset (ABODe)
ABODe is a University of Edinburgh DataShare dataset for behavior classification in group-housed mice using home-cage video, identities, bounding boxes, ground-plate positions, and annotator labels.
DANDI Archive for NWB datasets
DANDI is a BRAIN Initiative archive for publishing and sharing neurophysiology data, including electrophysiology, optophysiology, and behavioral data packaged as NWB and related standards.
International Brain Laboratory public data
The International Brain Laboratory public data releases expose standardized mouse decision-making experiments, including Neuropixels recordings, widefield calcium imaging, behavior, and session metadata accessed through the ONE API.
OpenNeuro
OpenNeuro is a free, open platform for sharing neuroimaging datasets, with public search, dataset pages, and download paths for web, S3, DataLad, and the OpenNeuro CLI.