小鼠自动化，定量认知/行为筛选：对于遗传学，药理学，动物认知与本科教学

The overall goal of the following experiment is to screen genetically and pharmacologically manipulated mice for malfunctions and basic mechanisms of cognition. This is achieved by making psychophysical measurements of physiological meaningful quantitative properties in a fully automated high throughput 24 7 behavioral phenotyping system. The next step is to analyze and graph the incoming data in quasi real time, which enables automated realtime control of testing on a mouse by mouse basis.

Next, the raw data, the test protocol code and all intermediate statistics are archived in a single data structure. In order to make the data publicly available and encourage further analysis, the results show strain specific alterations in the temporal vapor fraction based on the switch task. The main advantage of this technique over other methods, such as the water maze, is that it gives you physiologically meaningful results, very high throughput and no handling of the mice.

So this method can help answer key questions in the cognitive neurogenetics field, such as are there mutant strains that have Quantitatively altered interval timing mechanisms? The implications of this technique extend toward therapy or diagnosis of genetic cognitive deficits because it may lead to the discovery of genes that specify critical components of these mechanisms. Individuals new to this method need to understand the basics of MATLAB programming and to understand the power of quantitative automated Psychophysical testing.

Adam and I first had the idea for creating this system when we saw the labor intensive and un quantitative nature of the testing that was being done on genetically manipulated mice Set up the behavioral testing system in a standard laboratory room with 10 cabinets per room and eight test environments per cabinet. This enables 80 mice to be run at one time. Each test environment consists of a polypropylene mouse tub connected by an acrylic tube to a mouse test box.

The test box has three hoppers, which are monitored by infrared beams. The two outer hoppers have attached pellet feeders and can deliver food pellets while the middle hopper does not deliver food pellets. Cables passing through a port in a party wall, connect the test environments to the electrical interface cards and dedicated computers in another room, which run the protocol control programs.

One computer is required for every two cabinets. These computers are connected via a local area network to a server running the data analysis and graphing software. Set up the local area network so that the server on which data analysis software is installed can access the hard disks of the computers controlling the test environments.

Next, establish a file synchronizing account for data storage in the cloud. Add the TS system folder and its sub folders to MATLAB search path and be sure that the TS system folder is a cloud synchronized folder. The TS system is an open source software library of over 30 high level functions written in MATLAB that facilitate the creation of complex data analysis and data graphing code.

In the TS system toolbox call the function Ts begin to create the hierarchical data structure into which the raw data and all results derived from it will be placed by the other functions in the TS system toolbox. Next call ts add protocol to specify control parameters for an experimental protocol to specify decision code that will automate the decision to terminate the protocol and to go on to the next one and to specify the decision criteria to be used. Now, place the mice in the 24 7 live-in test environments, one mouse per environment.

Note the ID number of the mice, the box number and the letter that identifies the experiment control computer on the local area network and its IP address. Next call, ts start session to start the experimental session. Experimental sessions last one to two weeks during which time several different behavioral testing protocols are run.

Then at one control computer call the macro that will start all the boxes controlled by that computer. This macro was created by TS start session. Repeat this at the other control computers.

Data analysis can be done by writing new data analysis and graphing code using the commands in the TS system toolbox or by using or modifying existing data analysis and graphing code provided in the Ts library. Once begun, the experiment mostly runs itself, although some monitoring is required throughout the duration of the experiment. Monitor email for alerts from the TIA system data analyzing program indicating possible equipment malfunctions such as power failures, spontaneous control, computer reboots and pellet feeder malfunctions.

These email alerts allow a timely repair of the malfunction two to four times a day study. The plots of performance that the TS system data analyzing code produces every time it is called by the analysis timer. These plots can be monitored from anywhere in the world at any time, and if necessary, the experimental protocol can be revised online remote from the site where the mice are being tested.

Also use TS browser to study the data and summary statistics in the hierarchical data structure as they become available in quasi real time cumulative imbalance plots for pellets obtained referred to as income are plotted in red and visit duration is plotted in black for three wild type mice and three L one heterozygotes. A schedule reversal occurred at the point where the slopes turned downward marked by vertical dash red lines for three of the mice. The slope of the cumulative visit imbalance plot the black line closely tracks the slope of the cumulative income imbalance.

The red line, this indicates nearly perfect matching of the ratio of the average visit durations to the ratio of the incomes. Note how quickly the ratio of the visit durations adjusts to the change in the ratio of the incomes at the points of downward inflection. On the other hand, another mouse failed to match during the second feeding sessions as shown by the zero slope of the cumulative visit.

Duration imbalance plot in black, while the slope of the income imbalance in red is substantially positive. However, during the third feeding phase, the slopes are parallel, which means that this mouse began abruptly at the beginning of this phase to match its visit ratio to the income ratio. Clearly therefore, it was capable of doing so but did not do so for some reason during part of the testing.

This illustrates the importance of visualizing the entire course of performance rather than relying entirely on a few summary statistics. This mouse matched precisely during the first income ratio, but when it was reversed, the mouse did not fully reverse another mouse overmatched during the third feeding phase. That is its average visit imbalance was greater than its average income imbalance and it did not fully adjust to the reversal of the income imbalance.

Notice however that all six mice, both the three wild types and the three heterozygotes matched with precision during the very first four hour feeding phase and showed an abrupt response to the reversal of the income ratio. This demonstrates the intact functioning in the heterozygotes of many basic mechanisms of cognition, such as duration, estimation, number, estimation rate estimation, spatial localization, and the ability to remember simple abstract quantities. In cognitive neurogenetics, it is as important to determine what is still working normally as it is to find what is not working or has some interesting quantitative alteration in its operation.

Cumulative cycles or a visit to hopper one followed by a visit to hopper two, followed by a return to hopper one are plotted as a function of session time. Here the dashed green verticals mark the onsets of feeding phases. The solid red verticals mark.

The offset of feeding the rapidity with which the mouse begins to cycle between the hoppers is a measure of what might be called its boldness or tendency to explore. Five of the six showed some exploration before the onset of the first feeding phase and began abruptly to cycle between the hoppers as soon as it began. One mouse, however, did not show any explorative behavior at all, not a single cycle until midway through the second feeding phase, at which point it began abruptly to cycle between the hoppers.

Analysis of instrumental and classical conditioning experiments yield cumulative records of trial initiation speeds. These experiments measured the learning rate that is the number of pairings prior to the appearance of the conditioned behavior. The vertical dash lines mark the acquisition trial in each record.

The acquisition trial is the first trial on which there is a significant increase in trial initiation speed provided that the new higher speed is greater than the speed in the first segment. This ladder provision is required because sometimes there is a significant decrease prior to the first significant increase. Cumulative distributions of trial initiation latencies are shown here.

The point above the X axis at which the horizontal line at 0.5 intersects cumulative distribution is the median latency to initiate a trial. Cumulative records of the differences between the poking rate during the of a feeding hopper and the poking rate during the preceding inter trial interval are shown here. Initially, the slope of this record is zero or negative because the mouse pokes less during the hopper illumination than during the inter trial interval.

The slope turns positive when the mouse begins to poke into the illuminated hopper in anticipation of the impending delivery of appellate. The solid dots mark the point at which the slope becomes positive, which is the trial at which the conditioned response first reliably appears in these plots. The blue lines are empirical cumulative distribution functions showing the cumulative distribution of switch latencies for one mouse at three different settings of the feed latency at the long hopper.

The thin red curves that superimposed almost exactly on the empirical functions are the six parameter belga functions fit to these data. The heavy red lines are the corresponding probability density functions for the belga mixture distributions. Shortening the long feed latency caused the mouse to shorten the mean of its switch latencies and increase their precision.

It also caused the appearance of some impulsive switches as shown by the bump in the left tail of the probability density function in the bottom plot. The measure of this impulsivity is the fraction of the mixture attributed to the weibel component by the best fitting version of the Weibel Gauss mixture distribution. In this analysis, the short trials are marked with tiny red pluses at the bottom of the plot while the long trials are marked with tiny red pluses at the top of the plot.

The change in the density of the resulting red streaks at the bottom and top indicate the change in the relative frequency of the short and long trials. The switch latencies are marked by small open circles. Note that when the red at the bottom become more dense, meaning the short trials are more probable, the small circles shift upward and when the red at the top become denser, meaning an increase in long trial probability, the circles shift downward.

The blue lines are the medians of the distributions. A 24 hour raster plot of behavioral and environmental events over a 90 day test period is shown here. The black dots record pokes and the red and green dots record pellet deliveries at the two feeding hoppers.

The first black triangles record lights off while the second black triangles record lights on the first column of cyan triangles record the onset of the dusk feeding interval, which occurs one hour prior to dusk itself, and the second column of cyan triangles record the offset of the dusk feeding interval, which occurs three hours after lights off. These magenta triangles record the unsigned onset of the dawn feeding intervals, which occurs two hours before lights on, and the last column of magenta triangles record the offsets of the dawn feedings. The onsets of feeding anticipatory activity prior to the dusk feeding are marked with blue asterisks between six and 15 days.

After the beginning of testing, the mice begin to exhibit food anticipatory poking activity in the hour or so before the dusk feeding interval indicating the establishment of a circadian memory. Heterozygotes containing a mutation in the L one gene show greater interval timing precision than their wild type litter mate controls. This figure plots cumulative distributions of the precision estimates for the wild type controls as solid lines and their zygotic litter mates as dashed lines when run in blocks of trials with increasingly narrow intervals between the temporal goalposts, which is shown by the color differences.

In contrast, heterozygotes carrying the bat faced or BFC mutation show lower interval timing precision than their wild type ate controls. Once you've mastered this technique, the testing can be done in as little as one day or as much as a few weeks. Matching only takes one day food.

Anticipatory learning takes two or three weeks.