Garret Stuber, University of North Carolina at Chapel Hill, UNC Neuroscience center, Chapel Hill, USA
Joshua Matulonis, Research Foundation at Stony Brook University, USA
Prescilla Perrichon, IFREMER – La Rochelle – France
Nick Birch, Scottish Crop Researcher Institute, Dundee, UK
Marina Pleskacheva, Moscow State University, Lab. Of Physiology, Genetics and Behavior, Moscow, Russian Federation
Samuel Adeosun, University of Mississippi Medical Center, Jackson, USA
Thomas van Groen, University of Alabama at Birmingham, Cell Biology department, Brimingham, USA
Jacqueline Womersley, University of CapeTown, Department of Human Biology, Cape Town, South Africa
David Eilam, Tel-Aviv University, Zoology Department, Israel
Allan V. Kalueff, Department of Pharmacology, Tulane University Medical School, USA
Diane Lim, University of Pennsylvania, USA
Matt Andrzejewski, University of Wisconsin-Madison, USA
Joshua Herrington, Florida International University, USA
Craig Kinsley, University of Richmond, USA
Roman Bucher, University of Koblenz-Landau, Germany
Prof. Cachot, University of Bordeaux, France
EthoVision XT is powerful in data collection, but of course you will need to analyze this data. EthoVision XT helps you out in that department, too. It has powerful tools for data selection, visualization of your data, and analysis of your data.
In-depth analysis of your data
Visualizing your data – in data tables, plotted in graphs, and superimposed on your video file – is an important step in analysis. In fact, you can perform this before or after data selection, or even in the data acquiring process.
If you are interesting in certain behaviors such as body elongation or mobility, you can use the visualization function to fine-tune the thresholds for these behaviors.
Right after data acquisition, use visualization to get a good feel for your data. After data selection and analysis, visualization is a great way to illustrate certain results and present your data.
In addition to integrated visualization, a heatmap gives you an intuitive and unique view of your data. With one glance, you are able answer the question: in which part of the arena did the animal spend most of its time? You can select the data you want to appear in your heat map the same way you always select your data (see below). Use time intervals or nest different periods within a trial. Then sort a series of heatmaps in such a way that you can easily compare between trials, between individuals, or between groups.
You can also adjust your heatmap to your likings. Show the video image on the background, or not. Adjust the size of the blob and precision according to the size of your animal, or change the transparancy or the colors. Of course, you can easily export your heatmaps for use in presentations of scientific papers in all standard formats (PNG, JPG, BMP, GIF).
You can zoom in into tracks or parts of tracks by the powerful data selection criteria EthoVision XT offers. Selecting data in which animals received a specific treatment allows you to only analyze those data and ignore others. Or nest your data on time, zone or a behavioral state. The results can be visualized in a time-event plot together with the track data and the video file. This way you get an instant feel for your data. Analysis results are also represented in a table. Of course you can export your (raw) track data for further analysis with statistical software programs.
The number of parameters that can be used is substantial, including trial duration, in zone, distance to zone, distance to zone border, distance moved, velocity, heading, turn angle, angular velocity, mobility, proximity, rotations, elongation, etc. Statistics include mean, variance, standard deviation, standard error, minimum, maximum, sum, total, number of samples, etc.
For instance: use elongation, mobility, and head direction parameters when you are performing a novel object recognition test, a plus maze test, or a fear conditioning test. The stretch-attend posture, characterized by elongation of the animal’s body, is often considered a response to an object that incites fear or curiosity. Additionally, time spent immobile serves as a measure of fear. The rotation parameter is very useful in research on brain defects.
Of course you can export your (raw) track data for further analysis with statistical software programs.
In addition to batch acquisition, EthoVision XT also offers batch analysis and automatic analysis, which will save you a lot of time.
If you have large data files, it may take a while to process them. You can select which trial you want to subject to which data and analysis profiles beforehand and then EthoVision XT runs all the analyses sequentially. You can combine this with batch acquisition of your video files. Let EthoVision XT do the work while you concern yourself with other things - you don’t need to come back and start each separate track or analysis.
You can also compare the results between groups. For example, your experiment exists of two phases: training and testing. You are testing two different doses of a drug and of course, a control group. In total, that means you have six groups you might want to compare. EthoVision XT automatically groups the results. Data are displayed in tables and graphs, giving you an immediate feel for your results and making it easy to spot trends. Within these data, you can also fragment more using time bins or looking at the results per zone.
You can organize these graphs as you wish and adjust the design: you might want to compare treatment groups alongside or maybe you are interested in differences within treatment groups between training and testing. You can export your graphs in all standard formats (PNG, JPG, BMP, GIF).
Analysis of hardware events
If you are using the Trial & Hardware Control Module to control external equipment, EthoVision XT enables you to analyze these hardware events. This data can include signals that were sent out from EthoVision XT to external devices (such as to drop a pellet on turn on the house light) and signals that are sent back to EthoVision XT (such as the confirmation of a pellet drop or feedback from the running wheel). This helps you to answer questions such as: “How long was the house light on?” and “How much running did the mouse do?” in combination with the tracking data.
In any tracking system, there are three sources of noise that potentially affect the values of dependent variables such as distance moved or velocity; system noise, outliers, and small movements of the animal (body wobble). That is why you might want to smooth out your data. EthoVision XT has two methods for that. First is the Lowess filter, which fits a curve to the data set using a modified least square regression. The second method uses the defined ‘minimal distance moved’ to smooth out jitter and body wobble.