Data from: Precision cutaneous stimulation in freely moving mice
Data files
Mar 30, 2026 version files 13.53 MB
-
Fig2_DLC_LeftHindPaw.csv
4.05 MB
-
Fig2_DLC_TailBase.csv
5.71 MB
-
Fig2I_DLC_84R.csv
3.67 MB
-
Fig2K_HitAccuracy.csv
39.56 KB
-
Fig2L_LaserMAE.csv
33.59 KB
-
Fig5F_ActionRewardTimestamps.pkl
5.14 KB
-
Fig5H_AngleStd.csv
6.57 KB
-
Fig5H_CorridorSpeeds.csv
7.96 KB
-
README.md
7.98 KB
Abstract
Somatosensation connects animals to their immediate environment, shaping critical behaviors essential for adaptation, learning, and survival. Probing the relationships between somatosensory inputs and behavior in mice presents substantial challenges, primarily due to the practical difficulties of delivering stimuli to the skin during movement. To address this problem, we have developed a system for precise cutaneous stimulation of mice as they walk and run through environments. The system employs real-time body part tracking and targeted optical stimuli, offering precision while preserving the naturalistic context of the behaviors studied, thereby overcoming the traditional trade-offs between precision and animal behavior. We demonstrate the system from nociceptive testing conducted in standard small chambers to behavior in large complex environments, such as mazes. We observed that cutaneous inputs evoke rapid responses, which modify behavior when stimuli are applied during motion. This system provides a means to explore the diverse and integrative nature of somatosensation, from reflexes to decision-making, in naturalistic settings.
https://doi.org/10.5061/dryad.rr4xgxdhs
Corresponding Author Contact Information
Name: Liam E. Browne
Institution: University College London (UCL)
Email: liam.browne@ucl.ac.uk
First Author Contact Information
Name: Isobel Parkes
Institution: University College London (UCL)
Email: isobelp6@gmail.com
Dataset Overview:
This dataset contains files used for generating key figures, which demonstrate the development of closed-loop somatosensory stimulation in freely moving mice. Each .csv or .pkl The file corresponds to specific panels and figures in the manuscript, providing data for further analysis and validation.
Dataset File List:
1. Fig2_DLC_TailBase.csv
2. Fig2_DLC_LeftHindPaw.csv
3. Fig2K_HitAccuracy.csv
4. Fig2L_LaserMAE.csv
5. Fig2I_DLC_84R.csv
6. Fig5F_ActionRewardTimestamps.pkl
7. Fig5H_CorridorSpeeds.csv
8. Fig5H_AngleStd.csv
How to Use:
Each file corresponds to a particular figure or panel. Use the data in combination with the figures in the manuscript for reanalysis or extended exploration of behaviors in moving mice.
Funding:
This work was supported by a Sir Henry Dale Fellowship jointly funded by the Wellcome Trust and the Royal Society (109372/Z/15/Z to L.E.B.) and Medical Research Council (MR/N013867/1 to I.P.)
Ethics Approvals:
All animal work was carried out according to the UK Animals (Scientific Procedures Act (1986), approved by the UCL Animal Welfare and Ethical Review Body(AWERB), and performed under licenses issued by the UK Home Office.
Files and variables
File: Fig2_DLC_TailBase.csv
Description: Contains tail_base keypoint data with calculated locomotor speed categories (high/medium/low/stationary). This provides speed categories for Fig2_DLC_LeftHindPaw.csv used in Fig. 2H.
Variables
- x: estimated tail_base keypoint pixel x coordinate value
- y: estimated tail_base keypoint pixel y coordinate value
- likelihood: confidence in each tail_base keypoint estimation, expressed as likelihood
- dx: difference in the tail_base keypoint x coordinate value relative to the previous frame
- dy: difference in the tail_base keypoint y coordinate value relative to the previous frame
- dt: difference in the time relative to the previous frame, expressed in seconds
- distance: Euclidean distance for each frame, using dx and dy
- speed: smoothed speed, using Euclidean distance and dt
- category: locomotor speed category, determined by calculated smoothed tail_base keypoint speed
- mouse: unique identifier for each mouse
File: Fig2_DLC_LeftHindPaw.csv
Description: Contains left_hindpaw_mid keypoint data used to generate Fig. 2H, based on the locomotor speed categories from Fig. 2 _DLC_TailBase.csv.
Data Completeness Cells marked 'n/a' represent values that are not applicable. These values cannot be computed at the boundaries of the dataset. The first row (or frame) contains 'n/a' values for columns derived by calculating the change between consecutive frames, such as dx, dy, dt, distance, and speed. A rolling median filter with a window of 10 samples was applied to smooth the speed trace. This filter requires a window of 10 preceding data points to compute the speed value; the first 10 rows for the speed column are marked 'n/a'.
Variables
- x: estimated left_hindpaw_mid keypoint pixel x coordinate value
- y: estimated left_hindpaw_mid keypoint pixel y coordinate value
- likelihood: confidence in each left_hindpaw_mid keypoint estimation, expressed as likelihood
- dx: difference in the left_hindpaw_mid keypoint x coordinate value relative to the previous frame
- dy: difference in the left_hindpaw_mid keypoint y coordinate value relative to the previous frame
- dt: difference in the time relative to the previous frame, expressed in seconds
- distance: Euclidean distance for each frame, using dx and dy
- speed: smoothed speed, using Euclidean distance and dt
- speed_mm: smoothed speed, given in mm, using calibration factor of 0.45 pixels/mm
- mouse: unique identifier for each mouse
File: Fig2I_DLC_84R.csv
Description: Contains keypoints for the left and right paws along with locomotor speed and speed categories, for example, mouse 84R for plotting Fig. 2I.
Variables
- x: estimated keypoint pixel x coordinate value for the body part given in the body part column
- y: estimated keypoint pixel y coordinate value for the body part given in the body part column
- likelihood: confidence in each keypoint estimation, expressed as likelihood
- dx: difference in the keypoint x coordinate value relative to the previous frame
- dy: difference in the keypoint y coordinate value relative to the previous frame
- dt: difference in the time relative to the previous frame, expressed in seconds
- distance: Euclidean distance for each frame, using dx and dy
- speed: smoothed speed, using Euclidean distance and dt
- category: locomotor speed category, determined by calculated smoothed tail_base keypoint speed
- time: frame timestamp, given in seconds
- body part: body parts for which keypoint and metrics are computed
File: Fig2K_HitAccuracy.csv
Description: Contains hit/miss labels for different target body parts during stimulation to determine the accuracy of the system targeting during movement. Related to Fig. 2K.
Data Completeness Cells marked 'n/a' in the paw column represent values that are not applicable. On trials where the result was a miss, the laser did make contact with either paw, and therefore, the left or right paw is not recorded.
Variables
- frame: index of each stimulation frame analysed
- result: hit or miss label for each stimulation frame
- paw: identity of paw that was hit, whether targeted (left hind paw) or confused (right hind paw)
- mouse: unique identifier of the mouse
File: Fig2L_LaserMAE.csv
Description: Contains body part coordinates immediately 1 frame prior to stimulation (x_prestimulation, y_prestimulation) and laser spot coordinates on the following stimulation frame (x_stimulation, y_stimulation). This is used to calculate the MAE of stimulation from ground truth in Fig. 2L.
Variables
- mouse: unique identifier of the mouse
- body part: body parts for coordinates were determined
- category: locomotor speed category, determined by calculated smoothed tail_base keypoint speed
- x_prestimulation: ground truth pixel x coordinate value for the body part, a frame before stimulation
- y_prestimulation: ground truth pixel y coordinate value for the body par,t a frame before stimulation
- x_stimulation: ground truth laser spot x coordinate value on stimulation
- y_stimulation: ground truth laser spot y coordinate value on stimulation
File: Fig5F_ActionRewardTimestamps.pkl
Description: A .pkl file containing entry timestamps (entry_timestamps) and reward port trigger timestamps (reward_ts) for each corridor (right/left as suffix labels r/l) for four individual mice (m1, m6, m9, m13).
File: Fig5H_CorridorSpeeds.csv
Description: Contains mean speeds for each trial as mice moved through left and right corridors, shown in Fig. 5H.
Variables
- mouse: unique identifier of the mouse
- side: whether a particular trial was in the left or right corridor
- speed: mean smoothed tail_base keypoint speed for each trial as a mouse moves through the corridor
File: Fig5H_AngleStd.csv
Description: Contains the standard deviations of angles for each trial as mice moved through left and right corridors.
Variables
- mouse: unique identifier of the mouse
- side: whether a particular trial was in the left or right corridor
- angle: standard deviation of angles in each trial
