ADC information when subset of channels are saved

[alejoe91]

Follow up on SpikeInterface/spikeinterface#4144

Currently, we get the MUX information (which ADC each channel was recorded from) from the ProbeTable and we assign it to each contact as a contact annotation. This is problematic when the user saves a subset of channels: in this case, the contacts are sliced and the MUX table is also sliced. BUT, we need the full MUX table to compute inter-sample-shifts (eg.g, if each ADC samples 12 channels, after slicing the contacts we don't know which of the 12 channels we are left with).

A couple of possible solutions:

1. compute the inter-sample-shift directly here.
2. in addition to storing the MUX channels, we should also store the indices of the channels sampled within a cycle (and then compute the inter-sample shift from there)

@chrishalcrow @samuelgarcia @h-mayorquin what do you think?

[chrishalcrow]

Hello, I'd vote to compute something the inter-sample-shift directly here. It cannot change after the recording has occurred, so I think it's ok to fix it at the probe level.

[h-mayorquin]

I see that @alejoe91 went for version 2 (storing mux_index instead of computing inter-sample shifts directly in ProbeInterface).

I also vote for version 2. The reason is that I want ProbeInterface to focus on provenance information. Which contact belongs to which ADC and in what order within the ADC the contact is sampled is fundamental information about the probe hardware. Inter-sample shift, on the other hand, is an intermediate representation of this information used for the Fourier-based correction that we apply to compensate for the ADC multiplexing limitation. I prefer to store the former (hardware facts) rather than the latter (derived quantities).

I like what I see in #368, with the following remark:

I find mux_index and mux_channels confusing and somewhat clashing with other concepts, as "index" and "channel" are already overloaded terms in our ecosystem. Could we take this opportunity to improve the naming?

• For the index of the ADC to which the contact belongs, I propose adc_group or adc_index
• For the order in which the contact is sampled within the ADC, I propose adc_sample_position or adc_sample_order.

What do you think?

[alejoe91]

Great points @h-mayorquin

I agree that the proposed naming are much clearer! I also had the same reasoning about saving adc group and sampling order and not inter sample shift, which is more processing oriented.

[samuelgarcia]

My very first intuition was also 1. but I am OK with 2 and @h-mayorquin has a stronger arguments than intuition.