Repository navigation
Question about neuropixels channel mapping and signal names #39
Description
Hello,
We are in the process of setting up our neuropixels recording and I have 2 questions about the channel mapping for our NP1.0 probes:
- When we record in single-column mode, we end up with a channel map with 4 columns, but with no overlapping y-positions. It looks like the top half of the probe alternates between contacts on the two right-most columns (higher pair of x-positions) and then the bottom half alternates between contacts on the two left-most columns (lower pair of x-positions ).
I get the same result when I read the coordinates from the settings.xml file directly or when I use spikeinterface. The resulting map ends up looking something like this:
Is this correct? Because the single-column recording option suggests we can only expect a single column of contacts.
- Depending on the recording mode, the order of channel names in the xml file varies. Does this mean that the order of channels in the continuous.dat file also varies? Or is it simply the case that CH0 gives LFP0 and AP0, regardless of the order in the .xml file.
Thanks,
Susan
Activity
jsiegle commentedon Feb 11, 2025
This is correct. In this case, we consider the left two columns a single "staggered" column, and the right two columns another staggered column. Since the electrodes in each column are connected to the same set of output channels on each bank, the selected columns need to switch to the opposite side once you go from Bank A to Bank B.
The order that channel metadata gets saved to the XML file is based on their ordering along the shank, whereas in the .dat file the ordering is always based on the channel sequence. So the electrode connected to CH0 will always be the first one in the .dat file, followed by CH1, etc., even if this is not the case in the XML file.
SusanL82 commentedon Feb 13, 2025
Thanks, that helps a lot!