We're interested in understanding the mechanisms that regulate visual motion processing behaviors, such as the optokinetic reflux. We developed PyOKR as an accessible and uniform platform to reproducibly quantify a wide range of visual responses under various conditions to further our understanding of these behaviors. There's no unified method within the field to quantify optokinetic reflex responses, which can cause disparities when comparing data between laboratories.
We hope that PyOKR can help standardize how these data are analyzed to provide an accessible, unbiased, and robust tool to study these visual responses. Existing methods usually differ between laboratories and are often custom designed for their specific needs. PyOKR offers a unified method that is user-friendly, accessible, and adaptable to different experimental designs.
With a combination of automated analysis and user input, it can generate unbiased to accurate results to answer the user's desired questions. The use of our new PyOKR analysis method will facilitate the study of visual response behaviors in many contexts, such as genetic or pharmacological manipulation. Because of its accessibility and adaptability, it'll allow researchers to efficiently quantify responses to answer new questions about how visual circuits form and function.
With the help of PyOKR, we hope to identify new mechanisms that regulate the development of the direction-selective circuits that drive optokinetic reflexes. Using this approach, with neural circuit perturbation, will continue to help us study the development and function of these critical visual systems. To begin, download and install the required software for Python-based Optokinetic Reflex, or PyOKR, analysis.
On a Windows computer, run from PyOKR import OKR_win as o"followed by o.run. After opening the user interface, click File and then Open to open a browser for selecting the desired WAV file. To choose an output folder, click on Export folder.
Under the output file, enter the final analysis file name. Select Set Subject under File to set the program for an individual animal. To set stimulus parameters, go to Select Stimulus Direction and define a directionality among the four cardinal directions.
Under Select Stimulus Type, choose either unidirectional, oscillatory, or oblique. Then using, head and tail functions, set the length of time without stimulus at the beginning and end. Set the length of epoch, length of post stimulus, and number of epochs.
For unidirectional and oblique stimuli, set horizontal speed and vertical speed in degrees per second and specify the capture frame rate. For sinusoidal stimuli, adjust frequency and amplitude. Using Generate Stimulus Vector from Parameters, make the appropriate model from the input stimulus information and click on Select Epoch to scan through to total WAV file.
Click either Unfiltered Data or Filtered Data for preliminary adjustment to automatically select fast phase saccades based on maximal velocity changes. Under Unfiltered Data, confirm the saccades are accurately selected with a blue dot. Save the points with the middle mouse button and close the graph.
If automatic filtering is desired, set the filter Z-score threshold and click Filter Data to automatically filter saccades. After proper saccade selection, press Point Adjustment to select the region to remove. Alter the top and bottom points and save using the mouse buttons.
Using Set Polynomial Order, define the polynomial model that fits individual slow phases. Select Final Analysis to generate the slow phase models and calculate the distances, velocities, and tracking gains averaged across the epoch. Select View 2D Graph or View 3D Graph to view the two-dimensional or three-dimensional graph of the selected regions.
Select Add epoch to save to generate a collective values, and select View Current Dataset to view all added values and averages for a given animal. After repeating the entire process for all the files for a given animal, generate a final dataset containing all WAV data. Finally, export the data set via Export Data and proceed to the next animal data.
PyOKR analysis in Tbx5 conditional knockout mice indicated that these animals retain normal horizontal tracking gains as compared to wild type mice. However, these mice showed significant loss of vertical tracking with near zero gains in response to both upward and downward stimuli. Additionally, analysis of sinusoidal responses confirmed that Tbx5 conditional knockout mice exhibited greater horizontal tracking gains while showing significantly decreased vertical tracking.
