Behavioral Tracking and Neuromast Imaging of Mexican Cavefish

Cavefish is raising as a fantastic animal mural to understand evolution and biomedical processes. We are now conducting biomedical investigation by using these trait as a parallel of a human symptoms. Here we will show the battery of the behaviors assay and the analysis system that can be used for both evolutionary and biomedical studies.

To assay vibration attracts and behavior, a vibrating glass rod is inserted into a glass bowl containing one fish in the dark. Fish are acclimated four days prior to the assay in conditioned water with a 12, 12 light, dark cycle. The day prior to the day of the assay after three days of acclimation, replace water in assay chamber with fresh conditioned water.

On the day of the assay, fish are deprived of food until after the assay is completed, because cessation will change their response to vibrations. Prepare vibration-emitting apparatus by tuning to 40 Hertz. Set the proper recording parameters, stock conditions, and destination file.

Place assay cylinder on the recording stage, illuminated with infrared backlight in a dark room, and allow fish to acclimate for three minutes after being placed on the stage. After the three minute acclimation in the dark, record three minutes and 30 seconds of video. At the onset of the recording, insert the 7.5 millimeter diameter vibrating glass rod into the water column, approximately 5 centimeters in depth.

Avoid making any noise or vibrations while positioning the vibrating glass rod in the water, as fish can sense minor noises and vibrations. Make sure to finish this procedure within 30 seconds of starting the video recording to ensure that more than three minutes of behavior is recorded. Run batch file by double-clicking avs_creator.

bat to convert the compressed AVI video into readable format for ImageJ. The macro can be loaded by simply drag and drop into a graphical user interface shell. This macro will enable certain hotkeys for the following analysis.

In the working directory, create a new folder and name it Process_ImageJ. This folder will be automatically used to store output files from the following analysis. Right-click on the AVS video file to be analyzed and select quick mount.

After the AVS file is mounted as an external drive, open the AVI file in ImageJ. To set the scale of the distance measurement, select the diameter of the assay chamber by drawing a straight line across the chamber using the straight line selection tool, and then set the scale by clicking analyze, set scale function. For our case, we input 9.4 centimeters.

Check the radio box of global in order to standardize the scale across all the following video analysis. Copy the assay chamber area by using the oval selection tool, and then right-click and select image duplicate. At this time, specify the range of frames to keep for further analysis.

Clear outside of the assay chamber and convert the image to a binary image by hitting the hotkey seven on the number bar. After the background clears, you will be prompted to add a black dot at the center to indicate the position of the vibrating glass rod by using the oval selection tool already set to black with the fill function. Click OK and the macro will prompt you to move on to the threshold adjustment.

Adjust the threshold so that the fish is able to be seen throughout the entire video clip, then select apply. Run Tracker plug-in by hitting the hotkey eight on the number bar. Set the minimum pixel size to 100, when prompted.

This process will generate the distance between the vibrating glass rod and the fish per frame for all three minutes of the binary video. To save tracker file, hit the hotkey nine on the number bar. This will automatically export a binary stack of images of the entire video and an XLS file with coordinates and distance data, in case it is necessary to reanalyze.

This will also close all files associated with the current video. The files generated can now be further analyzed depending on the specific research question. Acclimate five experimental fish for four days or more in each chamber of a custom design 10 liter acrylic recording aquarium filled with conditioned water.

Each individual chamber is separated with black acrylic board. Each tank must be covered to prevent fish from jumping into another chamber. Set programmable power timer to automatically turn on white LED light for 12 hours and off for 12 hours everyday during acclimation period.

This will entrain fish's circadian rhythm if the fish is susceptible to entrainment. In order to provide a more diffused white light, we use a flat white opaque acrylic board of similar dimension to the 10 liter tank as a diffuser to pass light through for each tank. During this period, provide once a day feeding with live brine shrimp and provide aeration through sponge filter in each aquarium.

Note, ensure fish are fed at consistent time. For example, one times per day in the morning. The day prior to the day of the assay, three days or more of the acclimation, replace water in assay chamber with fresh conditioned water.

Set the recording the parameters in VirtualDub software. The video will be recorded in 15 frames per second in the fixed exposure time. Video is compressed in X264VFW codec to achieve approximately 700 times compression.

On the day of recording, feed each fish with live brine shrimp. Remove all sponge filter and start recording. Turn on the infrared light at the recording stage.

By observing VirtualDub live image on screen, adjust the position of each aquarium to make sure that they face the USB camera properly. Start 24-hour recording in the morning. For example, start time is 9 a.m.

and finish time is 9 a. m the following day. Start capturing the video and secure the location to avoid disturbance.

Occasionally check if the recording is running. After 24-hours recording, make sure the video saved correctly. Transfer the video to the PC workstation to track the fish's behavior.

After transferring the video to the appropriate folder, quickly do a quality check of the video to ensure it is of good quality. This includes looking at the lighting, checking if there's one fish in each section, and if there are any foreign movements that may cause mistracking. From here, the freeware SwisTrack is used for animal tracking.

The advantage of this software is that the software can track multiple animals at once and subtract background image by using an arbitrary frame we identify. The disadvantages are that tracking may jump between arenas containing a different fish. The background image may not be adequate for the later frames, and SwisTrack randomly assigns tracking ID to individual fish, therefore it is necessary to recover fish ID by plotting the average physical position of each fish.

To overcome these issues, three approaches are used. The use of binary masks, one for the tracking of even-numbered arenas, and one for the tracking of odd-numbered arenas. The use of Win-automation to automatically update background images every three minutes and the use of Perl scripts to reconstruct the fish IDs.

The first step of the analysis is to make the binary masks for odd and even arenas. After checking the video quality, choose a representative frame in a video editing software and use it to create two binary masks, one for even and one for odd-numbered arenas. The purpose of these binary masks is to avoid mistrackings between adjacent arenas.

After making binary masks, open the SwisTrack parameter files that determines the file paths and parameters for the tracking process. Prepare one parameter file in each of the even and odd file folders for the video tracking. In this file, type in the path to the file locations of the video and mask files.

After preparing the parameter files, open the Win-automation executable file that automatically resets the background every three minutes. With this executable file, SwisTrack will run automatically, then open the parameter file in each even or odd folder in this tracking software. Run the tracking software to track the movement of fish.

Within the first 9, 000 frames of tracking, ensure that the fish are being properly tracked by looking at the nearest neighbor tracking component. After propery establishing tracking, return to the adaptive background subtraction component tab and hit the hotkey R on the keyboard to run Win-automation for continuous adaptive background subtraction. After about five hours, text files are generated for each fish, containing X and Y coordinates for every frame of the video.

To continue the analysis, allocate the three Perl script files to the folder that includes both even and odd folders. In order to convert length in pixels to centimeters, it is necessary to know the centimeter to pixel ratio. Type this ratio in the appropriate location of the Perl script file named 1.fillupGaps.pl.

Run 1.fillupGaps. pl Perl script using a Unix emulator for Windows. First, use the CD command to move to the current working directory, then type Perl 1.fillupGaps.

pl to run Perl script. The other two scripts will be automatically called in sequential order. These three scripts assign each tracking file to each section of the aquarium, and thus can be used to analyze the behavior of each fish.

Confer with the text file named Summary_Sleep that is generated after running the scripts to double check that the number of frames dropped from the analysis is acceptably low. Missing fewer than 15%of frames is considered acceptable. Copy and paste tracking data into a spreadsheet file with an in-house macro.

This macro will restructure the analyzed data into several step process levels, such as average of sleep duration and sum of swimming distance. In order to stain fish lateral line, dissolve DASPEI stock solution in conditioned water for a final concentration of five micrograms per milliliter, then immerse the fish in the solution at room temperature in a dark environment for 45 minutes. After 45 minutes, recover the fish from DASPEI solution and immerse it in ice cold conditioned water with MES triple two.

Once the fish is deeply anesthetized, mount the fish in a Petri dish and observe it under efficient microscope. Make sure to take a statistic series to include all plans of the fish in focus and compile everything in a single picture. For image analysis, add a new folder entitled Process_ImageJ and the macro file to the folder containing the pictures to be analyzed.

Open ImageJ software and then open macro by dragging the macro file to ImageJ. Run macro by clicking macros, run macro. If the picture does not open automatically, click macro, file pickup.

Macro will then automatically pickup a picture file to analyze. For neuromast quantification shows the region of interest by clicking polygon two, and draw the area on the picture. Eight, five to duplicate region of interest.

Use paint tool to remove or add any dot that is extra or missing from the previous image and then eight, six. Once you eight, six, two new windows will appear. A window showing a scheme of numbered neuromast dots and a window with a table with total neuromast quantified.

Hit seven to save both files. Once you click OK, those files are stored and a new picture file will open for further analysis. Consolidate the neuromast count by running macro.

The following results are representative results of what can be expected for both morphs of fish under controlled conditions. These four figures show representative results from vibration attraction behavior assays. A and B are top views of the swimming path of A, a surface fish, and B a cavefish.

Red lines are the traces of the coordinates generated by the ImageJ tracking across three minutes of recording. C and D show a comparison of the results from C, a surface fish, and D, a cavefish, extracted by the spreadsheet macro from the TXT files. The Y-axis is the counted number of times the fish approached the vibrating rod.

Each dot represents one experimental observation during the behavioral assay of one fish. Note that surface fish do not show an increase in approaches at any frequency, while cavefish show a maximum of attraction around 35 to 40 Hertz. These two figures show representative results from sleep behavior assays extracted by the spreadsheet macros from the TXT files.

A and B demonstrate two of the metrics automatically summarized by the macro file. Figure A shows the swimming distance in daytime and nighttime for surface fish and cavefish, while B shows sleep duration for both morphs. In this case, cavefish showed hyperactivity and less sleep when compared to surface fish.

These figures show results of DASPEI image analysis extracted by the spreadsheet macro from the TXT files. A and B demonstrate the relationship between neuromast number versus vibration attraction behavior and neuromast size versus the behavior in cavefish, surface fish, and F1 progeny. Each dot represents each fish.

Note, the neuromast number is positively correlated with the vibration attraction behavior, suggesting that neuromast may regulate this behavior. Neuromast size also supports this conclusion. C and D are example images of DASPEI stained neuromast of C, a surface fish, and D, a cavefish.

Each fluorescent dot in the images is one neuromast, which has taken up DASPEI. Note, the noticeably greater abundance of neuromasts on cavefish. We showed a series of behavior analyses for lateral line-based behavior swimming activity and sleep.

The animal tracking system can also be adapted to other behaviors, such as stereotypic repetitive behaviors, social interactions, and behavioral laterality.