Tracy (Beta)

Note: Tracy is currently in beta — features and workflows may change regularly.

---

Table of Contents

1. Installation

2. Getting Started

   1. Launching Tracy
   2. Loading Your Movie
   3. Browsing Spots
   4. Generating Kymographs
   5. Generating Trajectories
   6. Browsing Trajectories

3. Additional Features

   • Drift Correction
   • Colocalization
   • Step Finding
   • Diffusion
   • Reference Image
   • Custom Columns
   • Color by Value

4. Saving & Loading

   • Save Trajectories
   • Load Trajectories
   • Import TrackMate Data

5. License

---

Installation

Install Tracy with your terminal:

# Create and activate a dedicated environment
conda create -n tracy python=3.10 -y
conda activate tracy

# Install Tracy
pip install tracyspot

Tip: On Windows, you can download Anaconda and use the Anaconda Prompt as your terminal.

---

Getting Started

1. Launching Tracy

conda activate tracy       # if not already active
pip install tracyspot --upgrade   # update during beta
tracy &                    # run in the background

Note: The first launch may take a few seconds.

2. Loading Your Movie

1. Click Load Movie (or use Load » Movie).
2. Select a single- or multi-channel TIFF movie.
3. If necessary, enter pixel size and frame interval when prompted.
4. Pan by holding down the middle button (or Ctrl/Cmd) and dragging, zoom with the mouse wheel.
5. If available, switch channels by clicking the channel label (shortcut: 1, 2, …).
6. Toggle the maximum projection with the button below the movie (shortcut: m).
7. Adjust the contrast using the slider.

3. Browsing Spots

1. Click on the movie to detect a spot at the click location.

   • Blue square: search region.
   • Magenta circle: fitted spot (radius = 2σ).

2. Hover over the inset to view a 3D fit (scroll to zoom, drag to rotate).

3. Hold r + scroll (or use Spot » Search Radius) to adjust the search radius.

4. Navigate frames with the slider under the movie.

The spot histogram shows intensities in the search area and highlights values in the spot.

4. Generating Kymographs

1. Switch to Line mode (slider under movie or n).
2. (Optional) Toggle max‑projection (m) to guide line placement or load a reference image.
3. Draw a segmented line by clicking anchors (press Esc to cancel).
4. Double‑click to finish and generate the kymograph.
5. For multi‑channel movies, a kymograph is generated for each channel (toggle with 1, 2, …).
6. Cycle through kymographs with , (previous) and . (next).

Tip: Apply a LoG filter via Kymograph » Apply LoG filter for clearer tracks (applies to subsequent kymographs).

5. Generating Trajectories

A. From a Kymograph

1. On the kymograph, draw a segmented line (blue anchors) to follow the track.
2. Double‑click to finish and compute a trajectory using your search radius and tracking mode (see tracking options).
3. Click any kymograph or plot point to jump to that spot; use →/← to step forward and backward.
4. Press spacebar to animate the trajectory.
5. If you want to recalculate the trajectory with new tracking options, press Enter (or Trajectory » Recalculate).
6. (Optional) Fill gaps via Kymograph » Connect Spot Gaps.

Trajectories are displayed on kymographs dynamically, so overlapping kymographs may share trajectories.

B. Direct Movie Tracking

1. Click an initial spot on the movie.
2. Skip a few frames and click the next spot. Repeat until the track is covered.
3. End a click sequence with Enter or cancel with Esc.
4. Browse/select tracks as above.
5. If you want to recalculate the trajectory with new tracking options, press Enter (or Trajectory » Recalculate).

You can use “video‑game” controls: w/a/s/d to move the cursor, l/j to change frames, k to select the spot.

Tracking Options

• Search Radius: adjust with r + scroll

• Tracking Mode (t):

  • Independent (default): fits each frame independently.
  • Tracked: uses previous frame’s spot as center.
  • Smooth: independent + post‑filter outliers.

• Tracking options are set for any subsequent analysis. An existing trajectory can be recalculated using the currently set options by pressing Enter (or Trajectory » Recalculate).

• If a spot looks wrong, you can invalidate it with x when it's highlighted.

• Avoid using spots in existing tracks via Spot » Avoid previous spots.

---

Plots

• Intensity Plot: integrated spot intensity at each frame.
• Speed Histogram: frame‑to‑frame speeds with the net speed overlayed (i.e. only considering start and end position).

---

Browsing Trajectories

• New trajectories append to the Trajectory Table.
• Click trajectories in the table or use the arrow keys (↑, ↓). Right‑click for options (e.g., Go to kymograph).
• Delete a trajectory with Backspace.

---

Additional Features

Drift Correction

1. Identify a stationary reference spot that can be found in most frames.
2. Click it and ensure it is found (i.e. a magenta circle appears). It does not matter which frame you choose.
3. Use Movie » Correct Drift to track it from start‑to‑end and apply the frame shifts.
4. Review and save if satisfactory.

Colocalization

• Determines colocalization if a spot exists within 4 pixels in the other channel.
• Toggle under Spot » Colocalization for multi‑channel movies.
• Existing trajectories prompt analysis
• Results appear as new table columns.

Step Finding

• Calculates steps in the intensity profile.

• Enable Trajectory » Calculate Steps.

• Adjust rolling‑average window and minimum step size:

  • Rolling average window: the smoothing window size (W) applied to the intensity trace before detecting steps. Larger values smooth noise more strongly but can blur short-lived steps; smaller values preserve fast changes but may be noisier. The window is in data points/frames (after invalid points are removed).
  • Minimum step size: the minimum intensity change required to accept a step edge. Increase this to ignore small/noisy fluctuations; decrease it to detect smaller steps. This threshold uses the same units as the intensity values shown in the plot.

• Existing trajectories prompt analysis.

• Results appear as steps in the Intensity Plot.

• Detected steps and sizes are saved in Per‑Trajectory sheet; each point’s step ID in Data Points.

Diffusion

• Estimates anomalous diffusion parameters from the trajectory’s mean-squared displacement (MSD):

  • MSD(Δt) = 4D · (Δt)^α (2D)

• Enable via Trajectory » Calculate Diffusion.

• Two analysis parameters control the MSD fit window:

  • Max lag: the largest time separation (Δt) included when computing MSD points and fitting D and α. Larger values include longer time scales but use fewer displacement pairs (and can be noisier).
  • Min pairs per lag: the minimum number of displacement pairs required to accept a given lag. If fewer pairs are available (e.g. short tracks or many invalid points), that lag is skipped.

• Requires pixel size and frame interval to be set (units: μm²/s for D, unitless for α). If either is missing, diffusion cannot be computed.

• Existing trajectories prompt analysis.

• Results appear as new trajectory table columns (e.g. D (μm²/s) and α).

• You can also Color By diffusion outputs (e.g. by α ranges) under Trajectories » Color By.

Reference Image

• Useful for overlaying filaments or guides during kymograph creation.
• Load via Load » Reference Image
• Toggle with the icon under the movie.
• While toggled, use shift+arrows to move the reference image if necessary

Custom Columns

• Right‑click any column header or use Trajectories » Add Column.

  • Binary: Yes/No flags.
  • Value: any numeric/text value.

• Assign via right‑click on table row or kymograph label.

Color by Value

• If custom column or colocalization data exists, under Trajectories » Color By choose binary, value, or colocalization.

---

Saving & Loading

Save Trajectories

• Save » Trajectories exports an Excel workbook with four sheets:

  1. Data Points: per-frame spot measurements along each trajectory.
  2. Per‑Trajectory: one-row summary statistics for each trajectory.
  3. Per‑Kymograph: aggregates grouped by kymograph/ROI (geometry).
  4. Aggregate analysis: a single-row summary across the whole movie.

---

Sheet: Data Points

Each row is one frame from one trajectory.

Columns

• Trajectory: trajectory ID.
• Channel: movie channel the trajectory was tracked in.
• Frame: 1-indexed frame number.
• Original Coordinate X / Y: the original (raw) coordinate for that frame.
• Search Center X / Y: the search center used for tracking in that frame.
• Spot Center X / Y: fitted spot center for that frame (blank if fit failed).
• Intensity: integrated spot intensity (blank if invalid/missing).
• Sigma: fitted spot σ (blank if fit failed).
• Peak: fitted peak amplitude (blank if fit failed).
• Background from trajectory: Yes if a fixed background was used for the trajectory, else No.
• Background: per-frame background estimate (blank if not computed).
• Speed (px/frame): frame-to-frame speed in pixels.
• Speed (μm/s) / Speed (μm/min): speed converted using pixel size + frame interval (blank if either is missing).

Optional columns: step finding

Only present if step-finding is enabled and steps exist.

• Step Number: step segment index for that frame.
• Step Intensity Value: median intensity for that step segment.
• Step Intensity Value (background-adjusted): step median minus the per-frame background (when available).

Optional columns: colocalization

Only present for multi-channel movies when colocalization is enabled.

• Colocalized w/any channel: Yes/No per frame (blank if not evaluated).
• Colocalized w/ch1, Colocalized w/ch2, …: Yes/No per frame for each channel (the reference channel column is left blank).

---

Sheet: Per‑Trajectory

Each row is one trajectory.

Columns

• Movie: movie file name.
• Trajectory: trajectory ID.
• Channel: channel the trajectory was tracked in.
• Start Frame / End Frame: 1-indexed start/end frames.
• Anchors: JSON list of anchor points used to define the kymograph track line (frame index + x/y in px; frame index is Tracy’s internal frame index).
• ROI: JSON description of the kymograph ROI geometry.
• Total Points: number of frames in the trajectory.
• Valid Points: number of frames with a valid intensity (>0).
• Percent Valid: 100 * Valid Points / Total Points.
• Search Center X Start / Y Start: starting search center coordinate.
• Search Center X End / Y End: ending search center coordinate.
• Distance (μm): straight-line displacement from start→end converted to μm (blank if pixel size missing).
• Time (s): duration from start→end in seconds (blank if frame interval missing).
• Background: fixed trajectory background value (blank if not used).
• Average Intensity: mean intensity over valid points.
• Median Intensity: median intensity over valid points.
• Net Speed (px/frame): straight-line displacement / (end-start frames).
• Net Speed (μm/s) / Net Speed (μm/min): net speed converted using pixel size + frame interval (blank if either is missing).
• Avg. Speed (px/frame): mean of per-frame speeds.
• Avg. Speed (μm/s) / Avg. Speed (μm/min): average speed converted (blank if either is missing).

Optional columns: step finding

Only present if step-finding is enabled and steps exist.

• Number of Steps
• Average Step Size: average absolute difference between consecutive step medians.
• Average Step Size w/Step to Background: as above, but also includes the final step-to-background difference when a fixed background exists.

Optional columns: diffusion

Only present if diffusion is enabled and could be computed (requires pixel size + frame interval).

• D (μm²/s): diffusion coefficient from MSD fit.
• α: anomalous diffusion exponent from MSD fit.

Optional columns: custom columns

Only present if you added custom columns in the UI.

• Custom columns appear as <Name> [binary] or <Name> [value] depending on the column type.

Optional columns: colocalization summary columns

Only present for multi-channel movies when colocalization columns exist.

• Ch. 1 co. %, Ch. 2 co. %, …: percent of frames colocalized with each other channel.
  • For 2-channel movies, the non-reference channel column is the overall colocalization percent.
  • For >2 channels, each non-reference column is computed separately per target channel.
  • The reference channel’s own Ch. X co. % cell is left blank.

---

Sheet: Per‑Kymograph

Each row is one ROI geometry (kymograph).

Columns

• ROI: ROI JSON (same format as in Per‑Trajectory).
• Total distance (μm): total polyline length of the ROI in μm (blank if pixel size missing).
• Total time (s): total movie duration in seconds (blank if frame interval missing).
• Number of trajectories: trajectories whose ROI matches this ROI JSON.
• Events (/min): Number of trajectories / (Total time in minutes) (blank if frame interval missing).
• Events (/μm/min): Events (/min) / Total distance (μm) (blank if pixel size or frame interval missing).
• Average net speed (μm/s): mean Net Speed (μm/s) across trajectories in this ROI.
• Average average speed (μm/s): mean Avg. Speed (μm/s) across trajectories in this ROI.
• Average run length (μm): mean Distance (μm) across trajectories in this ROI.
• Average run time (s): mean Time (s) across trajectories in this ROI.
• Average median intensity: mean Median Intensity across trajectories in this ROI.
• Average average intensity: mean Average Intensity across trajectories in this ROI.

---

Sheet: Aggregate analysis

A single row summarizing the whole movie.

Columns

• Pixel size (nm/px): pixel size used for unit conversions (blank if unknown).
• Frame time (ms): frame interval used for unit conversions (blank if unknown).
• Total movie frames
• Total time (s): total movie duration (blank if frame time unknown).
• Movie dimensions (px): width, height.
• Movie dimensions (μm): width, height converted using pixel size (blank if unknown).
• Total kymographs: number of ROI geometries (see multi-channel caveat below).
• Total kymograph distance (μm): sum of all ROI lengths (includes empty ROIs; blank if pixel size unknown).
• Empty kymographs: ROIs with zero trajectories.
• Number of trajectories
• Number of events (/min): total trajectories per total movie time in minutes (blank if frame time unknown).
• Number of events (/μm/min): events per minute divided by total kymograph distance (blank if pixel size or frame time unknown).
• Average net speed (μm/s), Average average speed (μm/s), Average run length (μm), Average run time (s), Average median intensity, Average average intensity: means across all trajectories, regardless of ROI.

---

Caveats (especially important for multi-channel movies)

• Per‑Kymograph and Aggregate “kymograph” counts are ROI-based, not per-channel. In multi-channel movies, Tracy draws a kymograph for each channel from the same ROI geometry, but the export currently groups by ROI geometry only. That means:
  • Total kymographs counts ROI geometries (not ROI × channels).
  • Per‑Kymograph rows are ROI geometries (not separate rows per channel).
• Overlapping ROIs / shared trajectories: if the same trajectory is associated with multiple ROIs/kymographs, the Per‑Kymograph sheet can double-count trajectories and bias the averages.

Load Trajectories

• Load .xlsx files with the above sheets or similar formats (requires columns: Trajectory, Channel, Frame, Search Center X, Search Center Y).
• Use Kymograph » Draw from Trajectories to redraw embedded lines.

Import TrackMate Data

• Load .csv from TrackMate (spot file) via Load Trajectories; uses TRACK_ID, FRAME, POSITION_X, POSITION_Y to perform a search.

---

License

This project is released under the MIT License — see LICENSE.txt.