An error in IBL-ONE

Software Support
,
1

Summary of what happened:

This following command
“sessions_with_spikes = one.search(project=‘brainwide’, data=‘spikes.times’)”
responded with this error “ValueError: Ambiguous search term “data””.

Could anyone give advice on how to solve this issue?

Version:

The latest versions of ONE and IBLLIB

Environment (Docker, Singularity / Apptainer, custom installation):

Windows, Anaconda, python 3.9

Data formatted according to a validatable standard? Please provide the output of the validator:

I did not use any validator, but I think it is standardized well.

2

Hello, we recently made changes to database (namely adding a dataset_qc search term) that makes data ambiguous. Instead, please use dataset.

3

Hello,

The search term that you are using is incorrect. You need to use dataset instead of data, like so

sessions_with_spikes = one.search(project='brainwide', dataset='spikes.times')

You can reference the documentation (Searching with ONE — ONE documentation) for more examples of searching. We also have this tutorial on getting started with ONE which has some useful examples and concepts.

Let us know if that resolves your issue.

Many thanks

4

Hi Mayo,

Thank you for answering it. The “dataset” option worked well.

Can I ask one more question here?

What is the sampling rate of spikes loaded by SpikeSortingLoader.load_spike_sorting()?
I could not make the values integer even by multiplying by 2500 or 1000.

Thanks.

5

Thank you for the answer as one of the developers.
We were able to run it well here.

May I ask one more question, the same one I asked Mayo?:

“What is the sampling rate of spikes loaded by SpikeSortingLoader.load_spike_sorting()?
I could not make the values integer even by multiplying by 2500 or 1000.”

6

Hello,

The spikes.times that are returned as part of the SpikeSortingLoader are the times of all of the spikes detected during the recording in chronological order. There is no fixed sampling frequency for these times.

If you are asking about the underlying ephys data from which these spikes were detected using spikesorting then the sampling frequency of the AP band is 30000Hz.

7

Thank you for your response.
I would like to change my question slightly.
Could you tell me the units of the spikes.times returned as part of the SpikeSortingLoader?
Are they in milliseconds?

Thank you.

8

Hello,

The units are in seconds from the start of the experiment.

9

Thanks.

So, regarding the output of spikes.times from the SpikeSortingLoader, the digits continue to very small decimal places. For example, if we want to split into 1ms bins, is it correct that we truncate the digits below 0.001?

10

I’m not sure you want to truncate, but you could bin all the spike that occur in those 1ms bins that you use.