AllenInstitute / ecephys_spike_sorting

Modules for processing extracellular electrophysiology data from Neuropixels probes
Other
109 stars 91 forks source link

Extracting NPX data recorded with Open Ephys on Linux #76

Closed vncntprvst closed 1 year ago

vncntprvst commented 1 year ago

This is sort of a follow-up on #61 : I was getting JSON file parameter validation errors, which I think are just due to the extract_from_npx_params fields npx_extractor_executable and npx_extractor_repo. The documentation on the extract_from_npx module says that NpxExtractor executable is available on Windows only but says nothing about the Github repository. There's no information on that in /extract_from_npx/_schemas.py or create_input_json.py either. There's also an indirect reference to that extractor on the Neuropix-PXI page. Where can this repository be found? Could you clarify the documentation on that topic? Is there any option to extract that data on Linux? The Open Ephys Python Tools package might be an option ?

jsiegle commented 1 year ago

Are you actually saving compressed NPX files directly from the Neuropixels Plugin? We added the ability to do this in an older version of Open Ephys because we were having trouble saving Neuropixels 3a data in the Binary format when running 3 instances of the GUI concurrently on the same machine. But if you're using 1.0 probes (with a PXI system) this should not be necessary (and is not even possible with the latest version).

If you're not using the NPX format, then you can just skip this module, which should fix the validation errors.

If you are using the NPX format, then you should decompress the files on the machine you used for data collection using the executable available here.

vncntprvst commented 1 year ago

Hi Josh, Thanks for the info. I do use 1.0 probes with a PXI system. I had other issues with subsequent modules (although not necessarily validation errors), so I assumed this was due to having by-passed the extraction module. The GUI documentation has no information on the compressed format - it's unclear if this is a default feature. As this is essentially a legacy steps, the documentation (and comments in the code) should be more explicit about it. I'll make suggestions in a PR. Thanks