OSCR

A transparent wheel-based platform for locomotion-on-demand and multi-view body and facial kinematics in head-fixed mice.

Code ↔ Paper

19 matches between paragraphs of the paper and lines of its authors' code, computed by the harvester (lexical-v1). Click a colored paragraph or line to see its counterpart.

The 19 matches · 1 of them tie a paragraph to a whole file, not to given lines: a weak match, whose lines are not tinted
  1. [1] § Methods › Data analysis › Video preprocessing ↔ Python/reduce_video_dimensions.py, lines 5–49 · score 0.81 · INTER_AREA, OpenCV, Resized videos, dimension, downscaling, quality
  2. [2] § Methods › Data analysis › Video preprocessing ↔ Python/notebooks/optical_flow.ipynb, lines 111–198 · score 0.79 · INTER_AREA, Resized videos, pupil video, optical flow, OpenCV, suffix
  3. [3] § Methods › Data analysis › Optical flow and motion-energy extraction ↔ Python/optical_flow.py, lines 125–202 · score 0.79 · avg_u, avg_v, optical flow, motion energy, OpenCV, timestamp
  4. [4] § Methods › Data analysis › Determination of eye/pupil area ↔ Matlab/load_eye_pupil_signal.m, the whole file · a weak match · score 0.77 · eye pupil roi, eye pupil signal, pupil area, largest, hole, calibration
  5. [5] § Methods › Data analysis › Optical flow and motion-energy extraction ↔ Python/src/utils/motion_analysis.py, lines 181–247 · score 0.75 · avg_u, avg_v, motion energy, OpenCV, timestamp, pose
  6. [6] § Results › Air-induced locomotion produces reliable paw kinematics on a transparent wheel ↔ Matlab/fig_dlc.m, lines 100–183 · score 0.71 · hind left, hind right, front left, front right, tail, nose
  7. [7] § Results › Air-induced locomotion produces reliable paw kinematics on a transparent wheel ↔ Matlab/fig_dlc_n3.m, lines 17–130 · score 0.70 · hind left, hind right, front left, front right, tail, nose
  8. [8] § Methods › Data analysis › LED intensity extraction for LED-based synchronization of video streams ↔ Matlab/fig_montage_n3.m, lines 257–387 · score 0.61 · nearest DAQ sample, video frame, divergence, windows, alignment, events
  9. [9] § Results › LED-based synchronization enables alignment of air stimulus across video streams ↔ Matlab/fig_montage_n3.m, lines 257–387 · score 0.60 · nearest DAQ sample, video frame, anchoring, divergence, windows, alignment
  10. [10] § Results › LED-based synchronization enables alignment of air stimulus across video streams ↔ Matlab/fig_montage.m, lines 135–172 · score 0.58 · paws camera, air state, video stream, montages, synchronization, alignment
  11. [11] § Results › Facial and eye dynamics during air-induced locomotion ↔ Matlab/fig_motion_energy_n3.m, lines 13–127 · score 0.56 · Optical flow, motion energy, LED signal, air onset, SEM, cm
  12. [12] § Methods › Data analysis › ROI selection and metadata ↔ Python/spyder_extract_led_signal.py, lines 13–86 · score 0.54 · interactive ROI, OpenCV, selection, frame, video
  13. [13] § Methods › Data analysis › Determination of eye/pupil area ↔ Matlab/fig_pupil.m, lines 17–94 · score 0.54 · pupil area, largest, hole, calibration, segmented, rectangular
  14. [14] § Results › Air-induced locomotion produces reliable paw kinematics on a transparent wheel ↔ Matlab/fig_motion_energy.m, lines 12–104 · score 0.54 · left axis, right axis, motion energy, box, offset, LED
  15. [15] § Results › Facial and eye dynamics during air-induced locomotion ↔ Matlab/fig_motion_energy.m, lines 12–104 · score 0.53 · left axis, right axis, motion energy, box, offset, LED
  16. [16] § Methods › Data analysis › ROI selection and metadata ↔ Python/fig_show_roi.py, lines 6–66 · score 0.53 · roi.json, OpenCV, selection, rectangular, pupil, frame
  17. [17] § Results › Air-induced locomotion produces reliable paw kinematics on a transparent wheel ↔ Matlab/fig_dlc.m, lines 59–98 · score 0.51 · Tracking reliability, Bar graphs, CDFs, cumulative, likelihood, frame
  18. [18] § Results › Air-induced locomotion produces reliable paw kinematics on a transparent wheel ↔ Matlab/fig_motion_energy_n3.m, lines 13–127 · score 0.51 · optical flow speeds, motion energy, air onset, SEM, paw, animals
  19. [19] § Results › Air-induced locomotion produces reliable paw kinematics on a transparent wheel ↔ Matlab/fig_dlc_F.m, lines 76–100 · score 0.50 · Tracking reliability, Bar graphs, CDFs, cumulative, likelihood, frame

Paper

Loaded from Europe PMC by your browser, not stored by OSCR: doi.org · Europe PMC

The paper is loaded when this pane is shown.

The authors' code

MATLAB · 673 lines · 21 KB · no license · 2 matches

  1. function fig_dlc
  3. % vp = evalin('base','vp');
  4. %
  5. % vp_l = evalin('base','vp_labeled');
  6. % vf = evalin('base','vf');
  7. % v = evalin('base','v');
  8. mD = evalin('base','mData'); colors = mD.colors; sigColor = mD.sigColor; axes_font_size = mD.axes_font_size;
  9. mData = mD;
  10. animal = evalin('base','animals(1)');
  11. % animal = evalin('base','fanimal');
  13. % physical_dist_cm = 5.0;
  14. % pixel_dist_px = 200; % Measure this from a still frame using 'imdistline'
  15. px_to_cm = 0.0114;
  17. n = 0;
  18. %%
  21. %% ---------- Load and Clean DLC Data ----------
  22. % Using the path structure you provided
  23. file_path = fullfile(animal(1).pdir,'video_20251216_165824DLC_resnet50_gcamp16declimbDec18shuffle1_185000_filtered.csv');
  25. % Set up import options to handle the triple-header (scorer, bodyparts, coords)
  26. opts = detectImportOptions(file_path);
  27. % Start reading from row 4 where the numeric data begins
  28. opts.DataLines = [4, Inf];
  29. opts.VariableNamingRule = 'preserve';
  30. tbl = readtable(file_path, opts);
  32. % Mapping likelihood columns based on your CSV image:
  33. % Column D=4 (Front Right), G=7 (Front Left), J=10 (Hind Right),
  34. % M=13 (Hind Left), P=16 (Tail Base), S=19 (Nose)
  35. lik_indices = [4, 7, 10, 13, 16, 19];
  36. bodyparts = {'Front Right', 'Front Left', 'Hind Right', 'Hind Left', 'Tail Base', 'Nose'};
  38. % Extract likelihood matrix [40748 frames x 6 bodyparts]
  39. likelihoods = table2array(tbl(:, lik_indices));
  41. %% ---------- Calculate Quality Metrics ----------
  42. p_thresh = 0.9;
  43. % Percentage of frames where p > 0.9 for each body part
  44. percent_high_conf = sum(likelihoods > p_thresh, 1) / size(likelihoods, 1) * 100;
  46. %% ---------- Updated Bodypart Colors (RGB) ----------
  47. % These values are sampled directly from the markers in your images:
  48. % Front Right (Blue/Indigo), Front Left (Sky Blue), Hind Right (Teal),
  49. % Hind Left (Bright Green), Tail Base (Orange), Nose (Red)
  50. custom_colors = [
  51. 0.20, 0.00, 1.00; % Front Right (Dark Blue/Indigo)
  52. 0.00, 0.60, 1.00; % Front Left (Sky Blue)
  53. 0.00, 0.80, 0.75; % Hind Right (Teal)
  54. 0.60, 1.00, 0.40; % Hind Left (Bright Green)
  55. 1.00, 0.60, 0.00; % Tail Base (Orange)
  56. 1.00, 0.00, 0.00 % Nose (Red)
  57. ];
  59. %% ---------- PANEL C: DLC Quality Analysis ----------
  60. magfac = mD.magfac;
  61. ff = makeFigureRowsCols(110, [3 5 4 1.5], 'RowsCols', [1 2], ...
  62. 'spaceRowsCols', [0.05 0.1], 'rightUpShifts', [0.1 0.1]);
  64. % --- C1: Bar Graph (% Frames > 0.9) ---
  65. subplot(1,2,1);
  66. hold on;
  67. for i = 1:6
  68. % Plot each bar individually to apply specific color
  69. bar(i, percent_high_conf(i), 'FaceColor', custom_colors(i,:), 'EdgeColor', 'none');
  70. end
  71. set(gca, 'XTick', 1:6, 'XTickLabel', bodyparts, 'XTickLabelRotation', 45);
  72. ylabel('% Frames (p > 0.9)');
  73. ht = title('DLC Tracking Reliability'); set(ht,'FontWeight','Normal');
  74. ylim([0 105]);
  75. box off;
  76. format_axes(gca);
  78. % --- C2: Likelihood CDF (Multi-line) ---
  79. subplot(1,2,2);
  80. hold on;
  81. for i = 1:6
  82. [f, x] = ecdf(likelihoods(:, i));
  83. plot(x, f, 'LineWidth', 2, 'Color', custom_colors(i,:));
  84. end
  86. % Reference line at 0.9 threshold
  87. line([p_thresh p_thresh], [0 1], 'Color', [0.5 0.5 0.5], 'LineStyle', '--', 'LineWidth', 1);
  89. xlabel('Likelihood');
  90. ylabel('Cumulative Probability');
  91. % legend(bodyparts, 'Location', 'northwest', 'FontSize', 7, 'Box', 'off');
  92. ht = title('Likelihood CDF'); set(ht,'FontWeight','Normal');
  93. grid on;
  94. box off;
  95. format_axes(gca);
  97. % Save to your designated PDF folder
  98. save_pdf(ff.hf, mD.pdf_folder, 'DLC_Quality_Analysis_Colored.pdf', 600);
  100. %% ---------- Setup & Data Loading ----------
  101. file_path = fullfile(animal(1).pdir, 'video_20251216_165824DLC_resnet50_gcamp16declimbDec18shuffle1_185000_filtered.csv');
  102. opts = detectImportOptions(file_path);
  103. opts.DataLines = [4, Inf];
  104. opts.VariableNamingRule = 'preserve';
  105. tbl = readtable(file_path, opts);
  107. % Define indices for X and Y based on your CSV structure
  108. % x: Col 2, 5, 8, 11, 14, 17 | y: Col 3, 6, 9, 12, 15, 18
  109. x_idx = [2, 5, 8, 11, 14, 17];
  110. y_idx = [3, 6, 9, 12, 15, 18];
  111. likelihood_idx = [4, 7, 10, 13, 16, 19];
  112. bodyparts = {'Front Right', 'Front Left', 'Hind Right', 'Hind Left', 'Tail Base', 'Nose'};
  115. % ---------- 1. Re-establish Time Base ----------
  116. N_sig = 40746;
  118. % If you have your sampling frequency (e.g., 30 fps or 60 fps)
  119. % If unknown, we can infer it if 'b.fs' exists from your earlier code
  120. if exist('b','var') && isfield(b,'fs')
  121. fs = b.fs;
  122. else
  123. fs = 60; % Defaulting to 30 fps; change this to your actual video FPS
  124. end
  125. fs = animal(1).video.specs.paws.fps;
  127. % Create time vector in seconds
  128. time_sec = (0:N_sig-1)' / fs;
  129. tmin = time_sec / 60; % Convert to minutes for plotting
  131. % Extract and truncate to match your signal time base (40745 samples)
  132. N_target = length(time_sec);
  133. X_coords = table2array(tbl(1:N_target, x_idx));
  134. Y_coords = table2array(tbl(1:N_target, y_idx));
  135. L = table2array(tbl(1:N_target, likelihood_idx));
  137. % Use the custom colors from the mouse markers
  138. custom_colors = [
  139. 0.20, 0.00, 1.00; % Front Right (Blue/Indigo)
  140. 0.00, 0.60, 1.00; % Front Left (Sky Blue)
  141. 0.00, 0.80, 0.75; % Hind Right (Teal)
  142. 0.60, 1.00, 0.40; % Hind Left (Bright Green)
  143. 1.00, 0.60, 0.00; % Tail Base (Orange)
  144. 1.00, 0.00, 0.00 % Nose (Red)
  145. ];
  147. % ---------- Likelihood filtering ----------
  148. lik_thresh = 0.9;
  150. low_conf = L < lik_thresh;
  152. X_coords(low_conf) = NaN;
  153. Y_coords(low_conf) = NaN;
  155. % ---------- Interpolate missing values ----------
  156. X_coords = fillmissing(X_coords,'linear');
  157. Y_coords = fillmissing(Y_coords,'linear');
  159. % ---------- No Smoothing ----------
  160. X_coordsS = X_coords;
  161. Y_coordsS = Y_coords;
  164. % ---------- Smooth coordinates ----------
  165. X_coordsS = movmedian(X_coords,5);
  166. Y_coordsS = movmedian(Y_coords,5);
  168. % X_coordsS = movmean(X_coordsS,5);
  169. % Y_coordsS = movmean(Y_coordsS,5);
  171. % Convert coordinates to cm
  172. X_cm = X_coordsS * px_to_cm;
  173. Y_cm = Y_coordsS * px_to_cm;
  174. fs = 60;
  175. % Compute velocity
  176. Vx = [zeros(1,6); diff(X_cm) * fs];
  177. Vy = [zeros(1,6); diff(Y_cm) * fs];
  179. % remove impossible spikes
  180. max_speed = 100; % cm/s
  182. Vx(abs(Vx) > max_speed) = NaN;
  183. Vy(abs(Vy) > max_speed) = NaN;
  186. %%
  187. T = animal(1).b.led_sig.paws;
  188. % t_paws = T.t_led/60;
  189. air_paws = double(T.is_on);
  190. %% ---------- Plotting: Trajectories vs Time ----------
  191. magfac = mD.magfac;
  192. tmin = time_sec / 60; % Time in minutes for the x-axis
  194. ff = makeFigureRowsCols(111, [3 5 6.9 1.5], 'RowsCols', [1 2], ...
  195. 'spaceRowsCols', [0.1 5], 'rightUpShifts', [0.1 0.2],'widthHeightAdjustment',[0 -300]);
  196. MY = 1920; ysp = 0.15285; mY = -2.5; titletxt = ''; ylabeltxt = {'PDF'}; % for all cells (vals) MY = 80
  197. stp = 0.35*magfac; widths = [3.12 3.12 2.85 1]*magfac; gap = 0.25*magfac;
  198. adjust_axes(ff,[mY MY],stp,widths,gap,{''});
  199. % axes_title_shifts_line = [0 0.55 0 0]; axes_title_shifts_text = [0.02 0.1 0 0]; xs_gaps = [1 2];
  202. % --- Subplot 1: X-Coordinates vs Time ---
  203. axes(ff.h_axes(1,1))
  204. hold on;
  205. for i = 1:6
  206. miny(i) = min(X_coords(:,i)); maxy(i) = max(X_coords(:,i));
  207. plot(tmin, X_coords(:,i), 'Color', custom_colors(i,:), 'LineWidth', 0.25);
  208. end
  209. % ylim([200 1400])
  210. ylims = ylim;
  211. plot(t_paws,air_paws*ylims(2),'k','LineWidth',0.25)
  212. ylabel('X Pixel Value');xlabel('Time (min)');
  213. % title('Horizontal Position (X) over Time');
  214. % legend(bodyparts, 'Location', 'eastoutside', 'FontSize', 7, 'Box', 'off');
  215. % box off; format_axes(gca);
  216. xlim([0 3]);
  217. onsets = find_rising_edge(air_paws,0.5,-1);
  218. offsets = find_falling_edge(air_paws,-0.5,1);
  219. ylims = ylim;
  220. [TLx TLy] = ds2nfu(time_sec(onsets(1)),ylims(2)-0);
  221. % axes(ff.h_axes(1,1));ylims = ylim;
  222. [BLx BLy] = ds2nfu(time_sec(onsets(1)),ylims(1));
  223. aH = (TLy - BLy);
  224. len = sum(find(time_sec(onsets)<3,1,'last'));
  225. for ii = 1:len%gth(onsets)
  226. [BRx BRy] = ds2nfu(time_sec(offsets(ii)),ylims(1));
  227. [BLx BLy] = ds2nfu(time_sec(onsets(ii)),ylims(1));
  228. aW = (BRx-BLx);
  229. annotation('rectangle',[BLx BLy aW aH],'facealpha',0.2,'linestyle','none','facecolor','k');
  230. end
  231. box off
  232. format_axes(gca)
  234. % --- Subplot 2: Y-Coordinates vs Time ---
  235. axes(ff.h_axes(1,2))
  236. hold on;
  237. for i = 1:6
  238. plot(tmin, Y_coords(:,i), 'Color', custom_colors(i,:), 'LineWidth', 0.25);
  239. end
  240. % ylim([10 1000])
  241. ylims = ylim;
  242. plot(t_paws,air_paws*ylims(2),'k','LineWidth',0.25)
  244. ylabel('Y Pixel Value');
  245. xlabel('Time (min)');
  246. % title('Vertical Position (Y) over Time');
  247. % box off; format_axes(gca);
  248. xlim([0 3]);
  250. onsets = find_rising_edge(air_paws,0.5,-1);
  251. offsets = find_falling_edge(air_paws,-0.5,1);
  253. ylims = ylim;
  254. [TLx TLy] = ds2nfu(time_sec(onsets(1)),ylims(2)-0);
  255. % axes(ff.h_axes(1,1));ylims = ylim;
  256. [BLx BLy] = ds2nfu(time_sec(onsets(1)),ylims(1));
  257. aH = (TLy - BLy);
  258. len = sum(find(time_sec(onsets)<3,1,'last'));
  259. for ii = 1:len%gth(onsets)
  260. [BRx BRy] = ds2nfu(time_sec(offsets(ii)),ylims(1));
  261. [BLx BLy] = ds2nfu(time_sec(onsets(ii)),ylims(1));
  262. aW = (BRx-BLx);
  263. annotation('rectangle',[BLx BLy aW aH],'facealpha',0.2,'linestyle','none','facecolor','k');
  264. end
  265. box off
  266. format_axes(gca)
  268. % Save the trajectory plot
  269. save_pdf(ff.hf, mD.pdf_folder, 'DLC_Trajectories.pdf', 600);
  271. %% ---------- Plotting: Inst Vel vs Time ----------
  272. magfac = mD.magfac;
  273. tmin = time_sec / 60; % Time in minutes for the x-axis
  275. ff = makeFigureRowsCols(111, [3 5 6.9 1.5], 'RowsCols', [1 2], ...
  276. 'spaceRowsCols', [0.1 0.061], 'rightUpShifts', [0.051 0.2],'widthHeightAdjustment',[-60 -300]);
  277. MY = 1920; ysp = 0.15285; mY = -2.5; titletxt = ''; ylabeltxt = {'PDF'}; % for all cells (vals) MY = 80
  278. stp = 0.1*magfac; widths = [3.12 3.12 2.85 1]*magfac; gap = 0.25*magfac;
  279. % adjust_axes(ff,[mY MY],stp,widths,gap,{''});
  280. % axes_title_shifts_line = [0 0.55 0 0]; axes_title_shifts_text = [0.02 0.1 0 0]; xs_gaps = [1 2];
  282. % --- Subplot 1: X-Coordinates vs Time ---
  283. axes(ff.h_axes(1,1))
  284. hold on;
  285. for i = 1:6
  286. miny(i) = min(Vx(:,i)); maxy(i) = max(Vx(:,i));
  287. plot(tmin, Vx(:,i), 'Color', custom_colors(i,:), 'LineWidth', 0.25);
  288. end
  289. ylim([min(miny) max(maxy)])
  290. ylims = ylim;
  291. % plot(t_paws,air_paws*ylims(2),'k','LineWidth',0.25)
  292. ylabel('Vx (cm/s)');xlabel('Time (min)');
  293. % title('Horizontal Position (X) over Time');
  294. % legend(bodyparts, 'Location', 'eastoutside', 'FontSize', 7, 'Box', 'off');
  295. % box off; format_axes(gca);
  296. xlim([0 3]);
  297. onsets = find_rising_edge(air_paws,0.5,-1);
  298. offsets = find_falling_edge(air_paws,-0.5,1);
  299. ylims = ylim;
  300. [TLx TLy] = ds2nfu(tmin(onsets(1)),ylims(2)-0);
  301. % axes(ff.h_axes(1,1));ylims = ylim;
  302. [BLx BLy] = ds2nfu(tmin(onsets(1)),ylims(1));
  303. aH = (TLy - BLy);
  304. len = sum(find(tmin(onsets)<3,1,'last'));
  305. for ii = 1:len%gth(onsets)
  306. [BRx BRy] = ds2nfu(tmin(offsets(ii)),ylims(1));
  307. [BLx BLy] = ds2nfu(tmin(onsets(ii)),ylims(1));
  308. aW = (BRx-BLx);
  309. annotation('rectangle',[BLx BLy aW aH],'facealpha',0.2,'linestyle','none','facecolor','k');
  310. end
  311. box off
  312. format_axes(gca)
  314. % --- Subplot 2: Y-Coordinates vs Time ---
  315. axes(ff.h_axes(1,2))
  316. hold on;
  317. for i = 1:6
  318. minyy(i) = min(Vy(:,i)); maxyy(i) = max(Vy(:,i));
  319. plot(tmin, Vy(:,i), 'Color', custom_colors(i,:), 'LineWidth', 0.25);
  320. end
  321. ylim([min(minyy) max(maxyy)])
  322. ylims = ylim;
  323. % plot(t_paws,air_paws*ylims(2),'k','LineWidth',0.25)
  325. ylabel('Vy (cm/s)');
  326. xlabel('Time (min)');
  327. % title('Vertical Position (Y) over Time');
  328. % box off; format_axes(gca);
  329. xlim([0 3]);
  331. onsets = find_rising_edge(air_paws,0.5,-1);
  332. offsets = find_falling_edge(air_paws,-0.5,1);
  334. ylims = ylim;
  335. [TLx TLy] = ds2nfu(tmin(onsets(1)),ylims(2)-0);
  336. % axes(ff.h_axes(1,1));ylims = ylim;
  337. [BLx BLy] = ds2nfu(tmin(onsets(1)),ylims(1));
  338. aH = (TLy - BLy);
  339. len = sum(find(tmin(onsets)<3,1,'last'));
  340. for ii = 1:len%gth(onsets)
  341. [BRx BRy] = ds2nfu(tmin(offsets(ii)),ylims(1));
  342. [BLx BLy] = ds2nfu(tmin(onsets(ii)),ylims(1));
  343. aW = (BRx-BLx);
  344. annotation('rectangle',[BLx BLy aW aH],'facealpha',0.2,'linestyle','none','facecolor','k');
  345. end
  346. box off
  347. format_axes(gca);
  349. % axes(ff.h_axes(1,1))
  350. % ylim([min([miny minyy]) max([maxy maxyy])])
  351. % axes(ff.h_axes(1,2))
  352. % ylim([min([miny minyy]) max([maxy maxyy])])
  356. % Save the trajectory plot
  357. save_pdf(ff.hf, mD.pdf_folder, 'DLC_Trajectories.pdf', 600);
  361. %% ---------- 1. Calculate Speeds for All Body Parts ----------
  362. % Ensure fs is defined (sampling rate)
  363. if ~exist('fs','var'), fs = 60; end
  364. dt = 1/fs;
  366. % Initialize speed matrix [Frames x 6 bodyparts]
  367. DLC_speeds = nan(size(X_coords));
  369. for i = 1:6
  370. % Calculate velocity (difference between frames)
  371. dx = diff(X_coords(:,i)) * fs; % Change in pixels per second
  372. dy = diff(Y_coords(:,i)) * fs;
  374. % Collapse X and Y into Speed (Magnitude of Velocity Vector)
  375. % We pad with a NaN at the start to maintain vector length
  376. DLC_speeds(2:end, i) = sqrt(dx.^2 + dy.^2);
  377. end
  379. % physical_dist_cm = 5.0;
  380. % pixel_dist_px = 200; % Measure this from a still frame using 'imdistline'
  381. px_to_cm = 0.0114;
  383. % Update the speeds by multiplying pixels by the calibration factor
  384. DLC_speeds_cm = DLC_speeds * px_to_cm;
  386. %% ---------- 2. Plot: Speeds vs Time (Panel F Style) ----------
  387. magfac = mD.magfac;
  388. tmin = time_sec / 60;
  389. magfac = mD.magfac;
  390. ff = makeFigureRowsCols(107,[3 5 6.5 1.5],'RowsCols',[1 1],'spaceRowsCols',[0.01 -0.02],'rightUpShifts',[0.2 0.22],'widthHeightAdjustment',[10 -250]);
  391. MY = 100; ysp = 0.15285; mY = -2.5; titletxt = ''; ylabeltxt = {'PDF'}; % for all cells (vals) MY = 80
  392. stp = 0.35*magfac; widths = [6.4 1 2.85 1]*magfac; gap = 0.115*magfac; adjust_axes(ff,[mY MY],stp,widths,gap,{''});
  393. axes_title_shifts_line = [0 0.55 0 0]; axes_title_shifts_text = [0.02 0.1 0 0]; xs_gaps = [1 2];
  395. % --- Subplot 1: Individual Speed Traces ---
  396. axes(ff.h_axes(1,1))
  397. hold on;
  398. for i = 1:6
  399. plot(tmin, DLC_speeds_cm(:,i), 'Color', custom_colors(i,:), 'LineWidth', 0.8);
  400. end
  401. ylabel('Speed (pixels/s)');
  402. title('DLC Speed: All Body Parts');
  403. % legend(bodyparts, 'Location', 'eastoutside', 'FontSize', 7, 'Box', 'off');
  404. box off; format_axes(gca);
  405. ylim([0 200]);xlim([0 3]);
  406. ylims = ylim;
  407. plot(t_paws,air_paws*ylims(2),'k')
  410. onsets = find_rising_edge(air_paws,0.5,-1);
  411. offsets = find_falling_edge(air_paws,-0.5,1);
  412. xlims = xlim;
  414. [TLx TLy] = ds2nfu(time_sec(onsets(1)),ylims(2)-0);
  415. % axes(ff.h_axes(1,1));ylims = ylim;
  416. [BLx BLy] = ds2nfu(time_sec(onsets(1)),ylims(1));
  417. aH = (TLy - BLy);
  418. len = sum(find(time_sec(onsets)<xlims(2),1,'last'));
  419. for ii = 1:len%gth(onsets)
  420. [BRx BRy] = ds2nfu(time_sec(offsets(ii)),ylims(1));
  421. [BLx BLy] = ds2nfu(time_sec(onsets(ii)),ylims(1));
  422. aW = (BRx-BLx);
  423. annotation('rectangle',[BLx BLy aW aH],'facealpha',0.2,'linestyle','none','facecolor','k');
  424. end
  425. box off
  426. format_axes(gca)
  428. save_pdf(ff.hf, mD.pdf_folder, 'DLC_Speeds_Combined.pdf', 600);
  430. %% ---------- 3. Statistical Comparison (Air-ON vs Air-OFF) ----------
  431. % Align with air epochs
  432. is_on = air_paws(1:length(avg_speed)) > 0;
  433. speed_on = avg_speed(is_on & isfinite(avg_speed));
  434. speed_off = avg_speed(~is_on & isfinite(avg_speed));
  436. if ~isempty(speed_on) && ~isempty(speed_off)
  437. [~, p_speed] = ttest2(speed_on, speed_off);
  438. fprintf('Global DLC Speed: Air-ON vs OFF, p = %.2e\n', p_speed);
  439. end
  442. %% ===================== DLC SPEED + AIR ON/OFF =====================
  444. % Inputs assumed from your script:
  445. % X_coords (N x 6), Y_coords (N x 6)
  446. % time_sec (N x 1) in seconds (DLC time base)
  447. % T = animal(1).b.led_sig.paws; T.time, T.is_on (LED time base)
  448. % bodyparts (1x6 cell), custom_colors (6x3)
  450. % ---- 0) Basic checks
  451. N = size(X_coords,1);
  452. assert(numel(time_sec)==N, 'time_sec length must match X_coords rows.');
  454. % ---- 1) Resample/align air signal to DLC time base
  455. % LED time is in seconds already (but you divided by 60 earlier for t_paws)
  456. % t_led = double(T.time(:)); % seconds
  457. t_led = double(T.t_led(:));
  458. air_led = double(strcmpi(string(T.is_on), "True") | double(T.is_on)); % robust
  460. % If T.is_on is already logical numeric, above still works.
  461. % Make sure time vectors are monotonic
  462. [~, idxSort] = sort(t_led);
  463. t_led = t_led(idxSort);
  464. air_led = air_led(idxSort);
  466. % Resample LED air to DLC time points
  467. air_dlc = interp1(t_led, air_led, time_sec, 'previous', 0);
  468. air_dlc(isnan(air_dlc)) = 0;
  469. air_dlc = air_dlc > 0.5;
  471. % ---- 2) Compute per-bodypart speed in pixels/s
  472. fs_dlc = 1/median(diff(time_sec)); % inferred FPS from DLC time base
  473. dx = [zeros(1,6); diff(X_coords,1,1)];
  474. dy = [zeros(1,6); diff(Y_coords,1,1)];
  475. speed_pix_s = sqrt(dx.^2 + dy.^2) * fs_dlc * 0.0114 ; % N x 6
  477. % Optional smoothing (helps readability)
  478. speed_pix_s_sm = movmedian(speed_pix_s, 5, 1);
  480. % ---- 3) Find air epochs on DLC time base
  481. air = air_dlc(:);
  482. on_idx = find(diff([0; air])== 1); % rising edges
  483. off_idx = find(diff([air; 0])==-1); % falling edges
  484. nTr = min(numel(on_idx), numel(off_idx));
  485. on_idx = on_idx(1:nTr);
  486. off_idx = off_idx(1:nTr);
  488. % sanity: ensure each onset precedes offset
  489. good = off_idx > on_idx;
  490. on_idx = on_idx(good);
  491. off_idx = off_idx(good);
  492. nTr = numel(on_idx);
  494. % ---- 4) Trial-wise paired comparison: Air-ON vs preceding Air-OFF window
  495. % Define a "pre" window immediately before each onset (e.g., 1.0 s)
  496. preWin_s = 1.0;
  497. preSamp = max(1, round(preWin_s * fs_dlc));
  499. mean_on = nan(nTr,6);
  500. mean_off = nan(nTr,6);
  502. for k = 1:nTr
  503. idx_on = on_idx(k):off_idx(k);
  504. idx_off = max(1, on_idx(k)-preSamp):on_idx(k)-1;
  506. if isempty(idx_off) || numel(idx_on) < 3
  507. continue
  508. end
  510. mean_on(k,:) = mean(speed_pix_s_sm(idx_on,:), 1, 'omitnan');
  511. mean_off(k,:) = mean(speed_pix_s_sm(idx_off,:), 1, 'omitnan');
  512. end
  514. % Drop trials with NaNs
  515. validTr = all(isfinite(mean_on),2) & all(isfinite(mean_off),2);
  516. mean_on = mean_on(validTr,:);
  517. mean_off = mean_off(validTr,:);
  518. nValid = size(mean_on,1);
  520. % Paired t-test per bodypart
  521. p_t = nan(1,6);
  522. tstat = nan(1,6);
  523. for i = 1:6
  524. [~, p_t(i), ~, stats] = ttest(mean_on(:,i), mean_off(:,i)); % paired
  525. tstat(i) = stats.tstat;
  526. end
  528. % ---- 5) Plot 1: Speed time-series with shaded air epochs
  529. figure(501); clf
  530. tmin = time_sec/60;
  532. for i = 1:6
  533. subplot(3,2,i); hold on
  535. % Shaded air windows
  536. yl = [0, max(speed_pix_s_sm(:,i), [], 'omitnan')*1.05 + eps];
  537. ylim(yl)
  539. for k = 1:nTr
  540. x1 = time_sec(on_idx(k))/60;
  541. x2 = time_sec(off_idx(k))/60;
  542. patch([x1 x2 x2 x1], [yl(1) yl(1) yl(2) yl(2)], ...
  543. 'k', 'FaceAlpha', 0.12, 'EdgeColor', 'none');
  544. end
  546. plot(tmin, speed_pix_s_sm(:,i), 'Color', custom_colors(i,:), 'LineWidth', 0.5);
  547. xlim([0 3]) % match your 3-min window
  548. xlabel('Time (min)')
  549. ylabel('Speed (px/s)')
  550. title(sprintf('%s', bodyparts{i}))
  551. box off
  552. end
  554. sgtitle('DLC point speed over time (shaded = Air ON)')
  556. % ---- 6) Plot 2: Onset-aligned speed (mean ± SD)
  557. win_pre = 2; % seconds before onset
  558. win_post = 5; % seconds after onset
  559. Lpre = round(win_pre * fs_dlc);
  560. Lpost = round(win_post * fs_dlc);
  561. t_rel = (-Lpre:Lpost)'/fs_dlc;
  563. % Build onset-aligned matrices: (time x trials x bodyparts)
  564. M = numel(t_rel);
  565. speed_onset = nan(M, nTr, 6);
  567. for k = 1:nTr
  568. c = on_idx(k);
  569. idx = (c-Lpre):(c+Lpost);
  570. if idx(1) < 1 || idx(end) > N
  571. continue
  572. end
  573. speed_onset(:,k,:) = speed_pix_s_sm(idx,:);
  574. end
  576. figure(502); clf
  577. for i = 1:6
  578. subplot(3,2,i); hold on
  579. X = squeeze(speed_onset(:,:,i)); % M x nTr
  580. mu = mean(X, 2, 'omitnan');
  581. sd = std(X, 0, 2, 'omitnan');
  583. plot(t_rel, mu, 'Color', custom_colors(i,:), 'LineWidth', 1.5);
  584. plot(t_rel, mu+sd, '--', 'Color', custom_colors(i,:), 'LineWidth', 0.75);
  585. plot(t_rel, mu-sd, '--', 'Color', custom_colors(i,:), 'LineWidth', 0.75);
  586. xline(0, 'k-');
  587. xlabel('Time from air onset (s)')
  588. ylabel('Speed (px/s)')
  589. title(bodyparts{i})
  590. box off
  591. end
  592. sgtitle('Onset-aligned DLC speed (mean ± SD)')
  594. % ---- 7) Plot 3: Trial-wise Air-ON vs Air-OFF (paired) bar + dots
  595. % figure(503); clf
  596. % tiledlayout(1,6,'Padding','compact','TileSpacing','tight');
  597. %%
  598. magfac = mD.magfac;
  599. ff = makeFigureRowsCols(107,[3 5 6.75 1.5],'RowsCols',[1 6],'spaceRowsCols',[0.01 0.04],'rightUpShifts',[0.051 0.2],...
  600. 'widthHeightAdjustment',[-45 -500]);
  603. for i = 1:6
  604. % subplot(1,6,i);
  605. axes(ff.h_axes(1,i));% nexttile
  606. hold on
  607. mOn = mean(mean_on(:,i), 'omitnan');
  608. mOff = mean(mean_off(:,i), 'omitnan');
  609. allmOn(i) = mOn;
  610. allmOff(i) = mOff;
  611. seOn = std(mean_on(:,i), 'omitnan');%/sqrt(nValid);
  612. seOff = std(mean_off(:,i), 'omitnan');%/sqrt(nValid);
  613. allseOn(i) = seOn;
  614. allseOff(i) = seOff;
  616. hb = bar([1 2], [mOff mOn]); % OFF then ON
  617. errorbar([1 2], [mOff mOn], [seOff seOn], 'k.', 'LineWidth', 1);
  618. set(hb,'FaceColor',custom_colors(i,:))
  619. % paired dots
  620. for k = 1:nValid
  621. plot([1 2], [mean_off(k,i) mean_on(k,i)], '-', 'Color', [0 0 0 0.15]);
  622. end
  624. set(gca,'XTick',[1 2],'XTickLabel',{'Air-OFF','Air-ON'});xtickangle(30)
  625. if i == 1
  626. ylabel('Mean speed (cm/s)')
  627. end
  628. % title(sprintf('%s | p=%.3g', bodyparts{i}, p_t(i)))
  629. ht = title(sprintf('%s | p<0.001', bodyparts{i})); set(ht,'FontWeight','Normal')
  630. box off
  631. format_axes(gca)
  632. end
  633. ht = sgtitle(sprintf('Paired Air-ON vs pre-Air-ON (Representative Animal, n = 34 trials)', nValid));set(ht,'FontSize',8,'FontWeight','Normal')
  635. save_pdf(gcf, mD.pdf_folder, 'DLC_bar_air_on_vs_off.pdf', 600);
  638. %% ---------- 1. Parameter Setup ----------
  639. win_pre = 5; % seconds
  640. win_post = 5; % seconds
  641. fs = 60; % Sampling rate (frames per second)
  642. Npre = round(win_pre * fs);
  643. Npost = round(win_post * fs);
  645. signal = DLC_speeds_cm(:,6); % Your data vector
  647. % Initialize arrays for mean values
  648. pre_means = [];
  649. post_means = [];
  651. % ---------- 2. Extraction Loop ----------
  652. for i = 1:length(onsets)
  653. idx = onsets(i);
  655. % Ensure the window is within the bounds of the signal
  656. if idx > Npre && idx + Npost <= length(signal)
  658. % Extract the pre-event segment (5s before onset)
  659. pre_segment = signal(idx - Npre : idx - 1);
  661. % Extract the post-event segment (5s starting at onset)
  662. post_segment = signal(idx : idx + Npost);
  664. % Calculate and store the mean for this specific trial
  665. pre_means(end+1) = mean(pre_segment, 'omitnan');
  666. post_means(end+1) = mean(post_segment, 'omitnan');
  668. end
  669. end
  671. % Display results for verification
  672. fprintf('Extracted means for %d valid trials.\n', length(pre_means));

fig_dlc.m at commit 5b9d1c9, no license · at the source

Overview

Authors: Pratik S. Paranjape1, Tahoura Mohammadi Ghohaki1, Samsoon Inayat1
  1. Department of Psychology, University of Nevada,Las Vegas, USA
Institutions: University of Nevada, Las Vegas (United States)
Journal: Scientific reports, volume 16, issue 1, article 26691
Dates: received 27 December 2025; accepted 4 May 2026; published online 21 May 2026
Type: Research article · Language: English
License: CC BY-NC-ND
Identifiers: DOI 10.1038/s41598-026-52322-z · PMID 42168450 · PMCID PMC13507272 · OpenAlex W7161960706
Open access: gold, a free copy (OpenAlex)
Status: code verified
Categories: other (modality), mouse (organism), methods / tools (subfield)
Methods: Connectivity, Statistics, fMRI & imaging, Physiology & signal measures
Keywords: Head-fixed locomotion, Stimulus-induced locomotion, Behavioral kinematics, Transparent running wheel, Multi-view videography, Optical flow, DeepLabCut, Mouse behavior, Biological techniques, Engineering, Neuroscience
MeSH: Face*, Locomotion*, Animals, Behavior, Animal, Biomechanical Phenomena, Male, Mice, Motion Capture, Video Recording (* major topic)
Topic: Zebrafish Biomedical Research Applications (Cell Biology, Biochemistry, Genetics and Molecular Biology), according to OpenAlex
Funding: Interdisciplinary Neuroscience Graduate Assistantship; Department of Psychology Graduate Assistantship; UNLV Internal Funding
Citations: not cited yet (Europe PMC); 32 references in the paper

Abstract

The abstract is not reproduced here: the paper's license (CC BY-NC-ND) does not allow it. Read it in the paper, at the publisher or on Europe PMC.

Repositories

Its files are read in the Code ↔ Paper reader above, with 19 matches between paragraphs and lines of code.

OSF k9j3b

License: none: the authors keep all their rights
State: the link answers, verified on 28 September 2026
Evidence: files inventoried
Size: 7 files, 0 scripts
Software Heritage: not checked
Found in: “Data availability”
Not found: README, license file, CITATION.cff, environment file, tests, continuous integration, documentation
Availability: 1 check, the latest on 28 September 2026: the link answers (HTTP 200)
  • 28 September 2026: the link answers (HTTP 200)
At the source: osf.io/k9j3b

neuromomentumlab/AIR_Wheel_Methods

License: none: the authors keep all their rights
State: the link answers, verified on 28 September 2026
Evidence: files inventoried
Commit: 5b9d1c94237197f3c21d7a9dd9459c1f5e15f62f, 1 May 2026
Languages: MATLAB (45), Python (15), Jupyter (4)
Size: 84 files, 64 scripts
Software Heritage: not archived
Found in: “Data availability”
Holds: 4 notebooks
Not found: README, license file, CITATION.cff, environment file, tests, continuous integration, documentation
Tools: OpenCV (10 files), Image Processing Toolbox (6 files), Statistics and Machine Learning Toolbox (5 files), NumPy (5 files), shadedErrorBar (5 files), Matplotlib (3 files), pandas (2 files), Signal Processing Toolbox (1 file), SciPy (1 file)
Availability: 1 check, the latest on 28 September 2026: the link answers
  • 28 September 2026: the link answers
64 files
At the source: github.com/neuromomentumlab/AIR_Wheel_Methods

The paper's code and data availability statement is in the Data section.

Tracing map

Proposed by the machine: these links were found in the paper and verified at the source, without human review. The map will receive a Zenodo DOI once one of the paper's authors has validated it with their ORCID.

What the map holds:

  • 2 repositories of the authors' code, each at its verified commit, with its license and how the link was found in the paper;
  • 64 scripts, each with its path and the digest of its content;
  • 19 matches between paragraphs of the paper and lines of the code (method lexical-v1);
  • neither the text of the paper nor the code itself.

Its JSON (tracing-map.json) is deposited on Zenodo with its DOI once the map is validated.

Data

No dataset and no data link were found in the paper.

Code and data availability statement

The paper has a code and data availability statement. Its license (CC BY-NC-ND) does not allow reproducing it here; in short, from what the harvester recognized in it:

Read it in the paper: doi.org/10.1038/s41598-026-52322-z.

Versions

The history of this record: each version stored by the harvester or made by a correction of its authors or of the maintainers of its code, and what changed in its facts. The texts of the paper (its abstract, its availability statements) are not part of it; versions that changed only those are not listed.

Version 1, 28 September 2026: the first record

Recorded: type, language, journal, volume, issue, pages, dates, 3 authors, 11 keywords, 9 MeSH terms, 3 funders, 32 references.

Cite

This paper

Paranjape, P. S., Ghohaki, T. M., & Inayat, S. (2026). A transparent wheel-based platform for locomotion-on-demand and multi-view body and facial kinematics in head-fixed mice. Scientific reports, 16(1), 26691. https://doi.org/10.1038/s41598-026-52322-z

BibTeX

@article{paranjape2026transparent,
author = {Paranjape, Pratik S. and Ghohaki, Tahoura Mohammadi and Inayat, Samsoon},
title = {{A transparent wheel-based platform for locomotion-on-demand and multi-view body and facial kinematics in head-fixed mice}},
journal = {Scientific reports},
year = {2026},
month = may,
volume = {16},
number = {1},
pages = {26691},
publisher = {Nature Publishing Group},
issn = {2045-2322},
doi = {10.1038/s41598-026-52322-z},
url = {https://doi.org/10.1038/s41598-026-52322-z},
pmid = {42168450},
pmcid = {PMC13507272}
}

RIS

TY - JOUR
AU - Paranjape, Pratik S.
AU - Ghohaki, Tahoura Mohammadi
AU - Inayat, Samsoon
TI - A transparent wheel-based platform for locomotion-on-demand and multi-view body and facial kinematics in head-fixed mice
T2 - Scientific reports
J2 - Sci Rep
PY - 2026
DA - 2026/05/21
VL - 16
IS - 1
SP - 26691
SN - 2045-2322
PB - Nature Publishing Group
DO - 10.1038/s41598-026-52322-z
UR - https://doi.org/10.1038/s41598-026-52322-z
LA - en
ER -

CSL-JSON

{
"id": "10.1038/s41598-026-52322-z",
"type": "article-journal",
"title": "A transparent wheel-based platform for locomotion-on-demand and multi-view body and facial kinematics in head-fixed mice",
"container-title": "Scientific reports",
"author": [
{
"family": "Paranjape",
"given": "Pratik S."
},
{
"family": "Ghohaki",
"given": "Tahoura Mohammadi"
},
{
"family": "Inayat",
"given": "Samsoon"
}
],
"container-title-short": "Sci Rep",
"volume": "16",
"issue": "1",
"page": "26691",
"DOI": "10.1038/s41598-026-52322-z",
"PMID": "42168450",
"PMCID": "PMC13507272",
"ISSN": "2045-2322",
"publisher": "Nature Publishing Group",
"URL": "https://doi.org/10.1038/s41598-026-52322-z",
"language": "en",
"issued": {
"date-parts": [
[
2026,
5,
21
]
]
}
}

The tracing map gets a citation of its own once an author has validated it and it has a DOI.

Similar papers

The papers with a page that share the most with this one: the tools found in their code, their categories, datasets, cited references and authors, the rarest counting most.

[1] doi:10.7554/elife.109717 [code]
Retrosplenial cortex enables context-dependent goal-directed sensorimotor transformation.
Journal: eLife
In common: OpenCV, Statistics and Machine Learning Toolbox, pandas, 3 other tools, mouse, 5 references
[2] doi:10.1038/s41467-026-75347-4 [code]
Sleep reveals dynamics integrating and segregating movement and stimulus representations in V1.
Journal: Nature communications
In common: Image Processing Toolbox, Signal Processing Toolbox, Statistics and Machine Learning Toolbox, 4 other tools, mouse, 3 references
[3] doi:10.1038/s41467-026-71725-0 [code]
Interactions across hemispheres in prefrontal cortex reflect global cognitive processing.
Journal: Nature communications
In common: Image Processing Toolbox, Signal Processing Toolbox, Statistics and Machine Learning Toolbox, 4 other tools, 3 references
[4] doi:10.1038/s41467-026-76581-6 [code]
Thalamocortical bursts encode reward contingencies and drive associative learning.
Journal: Nature communications
In common: OpenCV, Image Processing Toolbox, Signal Processing Toolbox, 5 other tools, mouse, 1 reference
[5] doi:10.1038/s41593-026-02232-0 [code]
Entorhinal cortex represents task-relevant remote locations independently of CA1.
Journal: Nature neuroscience
In common: OpenCV, Image Processing Toolbox, Signal Processing Toolbox, 5 other tools, mouse, 1 reference
[6] doi:10.1038/s41467-026-73476-4 [code]
Developmental molecular signatures define de novo cortico-brainstem circuit for skilled forelimb movement.
Journal: Nature communications
In common: OpenCV, Image Processing Toolbox, Signal Processing Toolbox, 3 other tools, mouse, 2 references
[7] doi:10.1016/j.neuron.2026.03.034 [code]
Dentate gyrus interneurons modulate winner-take-all network dynamics in freely behaving mice.
Journal: Neuron
In common: shadedErrorBar, Image Processing Toolbox, Signal Processing Toolbox, 5 other tools, mouse
[8] doi:10.1038/s41593-026-02255-7 [code]
Neural circuits encode prior knowledge of temporal statistics.
Journal: Nature neuroscience
In common: shadedErrorBar, Image Processing Toolbox, Signal Processing Toolbox, 5 other tools, mouse
[9] doi:10.7554/elife.108663 [code]
The Crunchometer, a low-cost, open-source acoustic analysis of feeding microstructure.
Journal: eLife
In common: OpenCV, Image Processing Toolbox, Signal Processing Toolbox, 5 other tools, methods / tools, mouse
[10] doi:10.1126/sciadv.adv5652 [code]
The anterior cingulate cortex modulates pupil-linked arousal.
Journal: Science advances
In common: OpenCV, pandas, SciPy, 2 other tools, 4 references

Contribute

The authors of this paper can claim it, correct its record and validate its tracing map, and the maintainers of its code (its owner, or a public member of its organization) correct what it says of their repository; anyone signed in can ask for its removal. Every request goes to OSCR's own machine, which answers it; your account page follows them.

Sign in with ORCID to claim this paper as one of its authors, correct its record or validate its tracing map: when the paper's metadata lists your ORCID iD, you are recognized at once. Maintainers of its code: sign in with GitHub, then claim the repository on your account page.

Request its removal

To ask OSCR to remove this record, the copies of its authors' scripts or its tracing map, use the removal request page: signed in, you say who you are, what to remove and why, then review and confirm the request. Published rules decide every request (how).

Discussion, reproductions, activity

Discussion: questions and error reports about this paper and its code, from signed-in readers and its authors. It opens with sign-in.

Reproductions: reports from readers who ran the authors' code: what they reproduced, with which environment, commit and data. It opens with sign-in.

Activity: what happens around this paper: new versions of its record, its map's validation, discussions and reproductions. It opens with sign-in.