In brief: Detailed settings for each probe.
On the Devices tab you checked the enable boxes for the probes you want to run and then clicked Detect.
Use this tab to configure parameters for each probe, individually.
A few settings are common to all the probes...
Select how the imec-provided ADC and gain calibration files for each probe are handled:
This mode runs the probe (and Obx ADC) sample fetching loop faster, which reduces the time to get fresh data. However, this drives the CPU 50%+ harder and reduces the maximum number of probes you can safely run concurrently.
Currently we are only supporting software-based initiation of imec data streaming. That is, the data acquisition starts manually when you click the Run button in the configuration dialog.
To do a run that is automatically customized for whole-probe surveys, check the box in this item group and select the number of seconds to record in each bank. The entire surface of each selected probe will be sampled: shank-0:[bank-0..bank-max], shank-1:[bank-0..bank-max], etc.
Although the survey controls the selected sites (looping over shanks and banks), it adopts the reference, gain, and highpass filter settings set in each probe's current IMRO table. That is, you can thus choose to do the survey with external or tip referencing by setting that choice for each probe in the IMRO editor.
A survey run will save data files to the current run directory specified on the Save tab. The files will automatically be named SvyPrb_date&time_g0_t0...
One (bin/meta) file-pair is saved for each probe. Within a file the sampling blocks are end-to-end. Open these files with the File Viewer (Ctrl-O) and open the Shank Viewer. As you scroll through the file the Shank Viewer shows the current probe region.
If enabled the sync waveform and/or the nidq stream are also recorded during a survey run. Using the linking feature of the File Viewer you can view probe activity with concurrent trigger events such as neural stimulations. These features enable you to program the most relevant regions of interest into the IMRO tables.
Note that the time to program a new bank plus amplifier settle time is about 2 to 3 seconds. You'll definitely see these transition regions as you browse the data.
In this mode all channels are saved, so the Save chans table column instead becomes Max bank. Use this item to skip sampling banks that are out of the brain.
As of SpikeGLX version 20230425, you can set up a filtered probe AP-band stream that runs in parallel with the raw stream. You can set the bandpass edges for this stream, and the stream automatically gets global demux CAR. These data, when present, are used to improve signal-to-background in both audio output and Shank Viewers (for {spike, AP pk-pk} calculations). There is a checkbox to disable this feature just in case you are running out of RAM or CPU, but we don't think this will be necessary unless you are running 12 probes or more at the same time.
SpikeGLX maintains a database of probe settings in the _Calibration folder. The settings are keyed to the probe's serial number. There is one entry per probe, hence, the software remembers last thing you did with that probe. The entries are each time-stamped. Individual entries are discarded after three years of disuse.
When you click Detect we look for each selected probe in the database. If found, the stored settings are used. If no entry is found, the probe gets default settings.
You can always see and edit those settings using the Each probe table on the IM Setup tab.
Upon clicking Verify or Run the table values are transferred to the database for future runs.
Configure dialog's outer edges or corners to resize all content.Default, or,For each probe, you can edit its:
More details below.
Not editable.
This is the probe's (headstage's) measured sample rate as determined by the calibration procedures on the Sync tab (See User Manual).
One checkbox per probe, but really the LEDs are on headstages...
When you click Verify or Run you will be warned if your LED settings for the same headstage conflict.
Use the graphical IMRO editor to select:
Edit directly in the table cell by specifying a channel list, just as you do for saved channels. E.g., 0:3,200,301:305.
The probe API lets you turn off the amplifiers for selected channels; what imec terms setting the channel to stand-by mode. The channel is still read out but its amplitude will effectively be zero. This makes graph windows easier to view. There is another more important way we use the bad channel list...
SpikeGLX and CatGT observe the geometry map that's stored in every recording's metadata file. A geometry (geom) map describes:
used in spatial averaging (CAR).When you click Verify or Run, SpikeGLX updates the probe's geom map. These channels are set to (used=0):
This lets you group/sort/order the channels in SpikeGLX graph windows. It has no effect on how binary data are stored.
You can save all of the channels being acquired by setting the list to any of:
You can save any arbitrary subset of channels using a printer-page-like list of individual channels and/or ranges, like: 1:200,205,207,350:360.
The SY channel is always added to your list because it carries error flags and the sync channel.
(If checked) AND (if this probe type has separate LF channels) THEN your list of saved channels is edited so that there is a saved LF channel for each listed AP channel. It's just a convenience.
The checkbox actually applies to all probes, not just the one being edited.
This tool allows you to alter the stored serial number and part number of the selected probe. It is only used for faulty EEPROM chips.
This is a handy way to set all of the editable fields:
Select a probe and copy its settings to a specified other probe, or to all other probes.
type.measured sample rates.bad chans.fin