Regarding channel mapping with Intan recording device

[kshtjkumar]

Hi, so i collected neural data from intan rhx device. We used two types of electrodes 1 is depth electrode(say channel number 28,29,30,31), other is a surface electrode (ECog, say channel number 1,3,5,7 ). Rest channels were deactivated. So now I just have 8 channel ids and i know based on channel number which channel id should correspond to what type of electrode. However when I upload the data using the extractor the channel ids are by default taken from '0' and I get an array of channel_ids = ['0', '1', '2'....]. This makes it difficult for me to group the channels based on type and process the data. Is it possible to see the original channel id as recorded by the intan device so that I can do the grouping easily?

[samuelgarcia]

Hi.
Noramly the id from the device (here intan) are preserved when you slice channels.
You could paste here simple code to check what your piepline please ?

[kshtjkumar]

import spikeinterface as si
file = '1st_DEC_LKO _MICE_Merged.rhs'
reader = si.extractors.neoextractors.intan.IntanRecordingExtractor(file,stream_name ='RHD2000 amplifier channel')
#reader = si.extractors.read_intan(file, stream_name ='RHD2000 amplifier channel')
recording_rhs = reader

recording_rhs.annotate(is_filtered = False)

channel_ids = recording_rhs.get_channel_ids()
fs = recording_rhs.get_sampling_frequency()
num_chan = recording_rhs.get_num_channels()
num_segments = recording_rhs.get_num_segments()

print("Channel_ids = ", channel_ids)
print("Sampling_frequency = ", fs)
print("Number of Channels = ", num_chan)
print("Number of segments = ", num_segments)

[kshtjkumar]

@samuelgarcia any solutions for the problem ?

[alejoe91]

hi @kshtjkumar

it is fixed in master ;)