Timing issue evident in PSTH

[bjhardcastle]

very obvious in 666986_2023-08-16 block 5:

• block 5: trials 368 - 458

• block 6: trials 459 - 552

• long frame interval (0.04s) in task hdf5 trial 421, frame 165707 (60% through block 5)

  • not exactly aligned with shift

• affects vis and aud stimuli

  • not a stim alignment issue (same issue when plot is aligned on trial start)

• affects all probes recorded

• sync processing otherwise appears normal, no warnings

• barcode intervals appear normal on sync and ephys

• does not affect lick times

• [ ] check drift map (currently broken https://github.com/AllenNeuralDynamics/aind-ephys-pipeline-kilosort25/issues/20)

[bjhardcastle]

This is a probe-sync issue (~200 ms of data dropped in ephys recording):

• method for getting sampling rate needs updating
• dropped periods need to be cut, and marked with obs_intervals in nwb