Chapter 5 - REANALYZING WAVEFORM DATA WITH THE LTP PROGRAM

Although this version of the program, LTP113E, does a reasonably good job of doing accurate on-line amplitude and slope analysis, experience in our group has indicated that subsequent off-line analysis can often improve the analysis accuracy. Doing Reanalysis with the LTP program is like reliving the experiment in ‘fast forward’ mode without any interruptions between ADsweeps.

To use the LTP program for off-line reanalysis, either repeat the procedure for going to MS-DOS Mode (Windows 95/98) or ‘Real’ Command Line DOS (Windows 3.x) as described above, or run directly from the Windows GUI. This can be achieved by going to the DOS compatibility box, going to the correct data directory and running the program. The only problem with reanalyzing in the DOS compatibility box is that there is less memory available for the LTP program and therefore fewer ADsamples/sweep that can be loaded and analyzed. For example, if LTP113E is run on a 32 MB Windows 95/98 computer in MS-DOS mode, 1,900,000 samples/sweep can be analyzed. However, if LPT101M is run on the same 32 MB Windows 95/98 computer in the DOS compatibility box, then only about 900,000 samples/sweep can be analyzed.

When analyzing ADsweep files taken with a Pico 42 board, make sure you start it up the program in reanalysis mode with this on the command line (just as you did when you were running the experiment, and even if running the reanalysis in the Windows DOS compatibility box) (see also Section 2.3.3):

5.1 Source of Pulse Stimulation Information for Reanalysis

First one has to choose whether the time of the start of the S0, S1 or Rm Pulse is taken from the disk file, or from the Pulse Stimulation Screen fields. Doing reanalysis in the same program that stimulates and acquires data has the added advantage of allowing one to use the Pulse Stimulation fields like a stimulus template and allows you to place the S0, S1 or Rm pulses at whatever time you want. To determine the source of the Pulse Stimulation information, use the menu command (Fig. 3.4.4A):

AmpFile -> Reanalyze suing ScreenField or DiskFile information

to choose the Choose Source of Reanalysis information dialog box (Fig. 3.4.4E) showing the following choices:

During on-line analysis the pulse time information contained in the Pulse Stimulation Screen Fields in the Pulse Stimulation Windows is used. During off-line Reanalysis, normally the pulse time information is obtained from the Disk File (the default).

However, if one wants to reanalyze a second synaptic potential of a paired pulse stimulation with LTP113E, one can choose the Source of Pulse Time Information for Reanalysis to be the Screen Fields and set the PrePulseDur field in the Pulse Stimulation Window (Fig. 3.1.2) to the time at which the second pulse occurs. The second synaptic potential will correctly be analyzed relative to the time set in the at the PrePulseDur field. (A later version of the LTP program will allow many S0 and S1 pulses to be analyzed simultaneously in one ADsweep.)

5.2 Automatically Clear Current Reanalysis at Start of Next Reanalysis

Sometimes you want to do then next reanalysis on the same calculation graph containing previously analyzed data points, and sometimes you want to do the next reanalysis on a new cleared calculation graph. This can be controlled by using the menu command (Fig. 3.4.4A):

AmpFile -> AutoClear Amp/Slope graphs for next reanalysis…

to call up and change the AutoClear Amp/Slope Dialog Box (Fig. 3.4.4D). This dialog box contains :

If Yes is chosen, then doing the next reanalysis (Section 5.4) will not only clear the Amp/Slope Calculation graphs, but will also start new Amplitude/Calculation (*.AM0 and *.AM1) files.

If No is chosen, then doing the next reanalysis Amp/Slope calculation points will be superimposed on the points already present on the Amp/Slope Calculation graphs, and the new data points will be appended to the present Amplitude/Slope calculation (*.AM0 and *.AM1) files.

5.3 Digital Filtering

The current version of the LTP program, LTP113E, cannot do on-line data acquisition, signal averaging and digital filtering at the same time (a subsequent version will be able to do this). Because of this, and because digital filtering is slow, it has tended to be used during reanalysis.

Digital Filtering can be very useful for reanalyzing Peak Amplitude because the algorithm to measure peak amplitude merely picks out the most positive or negative ADsample. Therefore, if there is a lot of noise, the Peak Amplitude measurement will be artificially increased by the extraneous noise. Appropriate levels of digital filtering can reduce this noise and give a more accurate Peak Amplitude measurement. (A later version of the LTP program will do a third-order polynomial curve fit (a skewed parabola) to the peak, so digital filtering will not be necessary for accurate Peak Amplitude determination.)

Digital filtering shouldn’t be necessary for analyzing the Slope since the slope is a linear regression line (least squares fit) through the data points. Interestingly, digital filtering can reduce the sweep-to-sweep jitter in the slope data points, possibly because the digital filtering effectively extends the time over which the slope is calculated by including sample points before and after the time at which the slope is calculated.

Section 4.2 shows how to turn on and use the digital filtering. Fig. 4.2.1 shows Filter cutoff freq field which changes the amount of digital filtering. When digital filtering is on, the amplitude/slope calculation analysis is done on the filtered ADsweeps only.

5.4 Reanalysis

To reanalyze data use the pulldown menus and dialog boxes to choose those files you wish to reanalyze. Remember, every data file is 1 sweep (or 1 averaged sweep), so you can select only those sweeps you wish to reanalyze. To do this enter the menu command (Fig. 3.4.4A):

AmpFile -> Reanalyze AD sweep files

to enter the Reanalysis Dialog Box (Fig. 3.4.4B). Then:

The analysis of the chosen ADsweep files will automatically go at an interval set by the

field value set in Miscellaneous Values Window (Fig. 3.1.6).

Once analysis has begun, it can be temporarily halted anytime by pressing the SPACE bar. The detection values can be changed and that will change the calculation of the slope in the current ADsweep file. The ADsweep graph values can also be changed, and different Windows can be chosen.

Pressing the SPACE bar again will move to the next ADsweep so this will allow single stepping through all the ADsweeps and customizing each slope detection if need be. (However, it is less biased to have one detection setting for reanalyzing the whole experiment.)

Pressing F3 will resume or Continue the reanalysis at the normal Sweep Repeat Period.

Pressing F4 will Stop the Reanalysis.

Remember that during a reanalysis run, changing the detection parameters will change the amplitude/slope of last analyzed ADsweep, and will change that value in the Amp/Slope (*.AM?) file. Therefore if you are changing detection parameters ‘between’ reanalysis runs, do so on the first sweep of the next reanalysis run and not on the last ADsweep of the previous reanalysis run (e.g. supposedly ‘between’ reanalysis runs).

Also remember to increase the amplitude sensitivity as much as possible, and set AutoReset Timebase to Off to be able to zoom into the appropriate waveform times for a more accurate reanalysis (see Section 4.7).

Go to LTP Manual - Chapter 6

Return to the LTP Manual (Table of Contents)

Return to the LTP Home Page