Two issues about parameters in loading data

55 views
Skip to first unread message

Junying Su

unread,
Feb 13, 2025, 1:54:35 AMFeb 13
to Open Ephys

Dear OpenEphys developers, 

I have two issues about loading the ONIX data files.

https://open-ephys.github.io/bonsai-onix1-docs/articles/hardware/np1e/load-data.html

The first issue:

ap_recording = se.read_binary(os.path.join(data_directory, f"np1-spike_{suffix}.raw"), 3e5, np.uint16, num_channels, gain_to_uV=ap_scalar, offset_to_uV=-ap_offset)
lfp_recording = se.read_binary(os.path.join(data_directory, f"np1-lfp_{suffix}.raw"), 3e5/12, np.uint16, num_channels, gain_to_uV=lfp_scalar, offset_to_uV=-lfp_offset)

that is, ap sampling frequency = 300,000 Hz, lfp sampling frequency = 25,000 Hz. However, in Neuropixels 1.0 User Manual shown in the picture, ap sampling frequency = 30,000 Hz, lfp sampling frequency = 2,500 Hz. Why are these different?

The seond issue:

What is the meaning of 1.2e6? Is it related to the input voltage range in Neuropixels 1.0 User Manual?

Thank you.

Junying Su

unread,
Feb 13, 2025, 1:55:51 AMFeb 13
to Open Ephys
图片1.jpg

Junying Su

unread,
Feb 13, 2025, 2:13:37 AMFeb 13
to Open Ephys
An additional issue:
When recording data, the AP Gain and the LFP Gain are set to 1000 and 50, respectively. However, when loading data, ap_gain = 500 and lfp_gain = 500. Why are these different?
Thank you.

Junying Su

unread,
Feb 13, 2025, 2:16:32 AMFeb 13
to Open Ephys

Josh Siegle

unread,
Feb 21, 2025, 7:26:36 PMFeb 21
to Junying Su, Open Ephys
Hi Junying,

There's a mistake in the example code – the frequency values should be 3e4 and 3e4/12. I just opened an issue to correct this here: https://github.com/open-ephys/bonsai-onix1-docs/issues/214

You're correct that the 1.2e6 refers to the input range, which is ±0.6 V peak-to-peak, or 1.2e6 microvolts spanned by 10 bits of the ADC. I think the Neuropixels datasheet may be wrong here, I'm not sure where the ±5 mV value comes from.

For the AP and LFP gain settings, you'll need to substitute the values that were actually used during the recording.

Josh


--
You received this message because you are subscribed to the Google Groups "Open Ephys" group.
To unsubscribe from this group and stop receiving emails from it, send an email to open-ephys+unsubscribe@googlegroups.com.
To view this discussion visit https://groups.google.com/d/msgid/open-ephys/645daa6f-5e86-47f0-9093-9f0ac9334e2an%40googlegroups.com.

Reply all
Reply to author
Forward
0 new messages