Converting from LabChart7 Intracellular Electrophysiology Data

[mahironoda]

Hello NWB Community!

I’m a new user and looking for some help converting intracellular electrophysiology recordings from single muscle fibers (mice & Drosophila).
We use LabChart7 to record the data. While NWB doesn’t support direct conversion from the LabChart data file (.adicht), LabChart can export the data into a text file, European Data Format, etc.
Our end goal is to be able to save the recording data (membrane potential, EPPs/mEPPs) along with metadata (animal strain, sex, DOB, age; electrode resistance, stimulation frequency, etc). The NWB file will eventually be uploaded onto the DANDI archive.

If anyone here has experience using NWB with a similar type of data and is willing to describe their workflow/conversion steps, or have a quick conversation with me about it, please let me know. I can be contacted at nodamahi [at] grinnell [dot] edu.

Thank you!

[yarikoptic]

@CodyCBakerPhD @bendichter have you had fun with such files?

[bendichter]

I don't believe we have. Do you have a user in mind or any example data?

[mahironoda]

Here is an example data file (text file export of LabChart 7 data). Regarding users, we will be using NWB as a format to archive and share our lab data as part of the NIH grant requirements. We will upload the NWB datasets to the DANDI archive.

[mahironoda]

Here is an example data file (text file export of LabChart 7 data). Regarding users, we will be using NWB as a format to archive and share our lab data as part of the NIH grant requirements. We will upload the NWB datasets to the DANDI archive.

[kabilar]

Hi @CodyCBakerPhD, do you have any suggestions regarding this question? Thanks.

[CodyCBakerPhD]

@mahironoda There are two approaches:

1. Ask the nice people over at NeuroConv if they can add this format to their Icephys suite ('New Issue' -> 'Request Format Support') - this will probably involve adding NEO support as well

2. Put together a PyNWB/MatNWB conversion script and parse it yourself

Here is a small starter script I put together to extract the main file content (per segment)

import pathlib

import numpy

# Lab Chart 7 text (mixed TSV) format
file_path = pathlib.Path(".../03_31_2025.txt")
content = lab_chart_7_file_path.read_text()
lines = content.splitlines()

segment_counter = -1
segment_string = ""
parsed_lines_per_segment: dict[str, list[list[float, float]]] = dict()
for index, line in enumerate(lines):
    entries = line.split("\t")

    if entries[0] == "Interval=":
        segment_counter += 1
        segment_string = f"seg{segment_counter}"

        parsed_lines_per_segment[segment_string] = []
    elif "=" in entries[0]:
        continue

    if len(entries) == 3 and entries[2] != "-":
        try:
            parsed_lines_per_segment[segment_string].append([float(entries[0]), float(entries[2])])
        except:
            print(entries)

timestamps_per_segment = {
    segment: numpy.array([values[0] for values in parsed_lines], dtype=float)
    for segment, parsed_lines in parsed_lines_per_segment.items()
}
data_per_segment = {
    segment: numpy.array([values[1] for values in parsed_lines], dtype=float)
    for segment, parsed_lines in parsed_lines_per_segment.items()
}

The next part of which would be to follow the PyNWB Icephys Tutorials to create all of the rich icephys metadata and then write it all to an NWB file

After that, simply follow the main DANDI instructions for creating a new Dandiset and uploading your content

[mahironoda]

@CodyCBakerPhD Thank you very much for these thorough suggestions! I will try these steps and will start a new thread if I encounter specific issues. Thank you again for your time!

[CodyCBakerPhD]

@mahironoda No problemo, happy to help!

Oh, do note that that script I included skips rows that end in that weird #* -t pattern. Not sure what that means or if the data should be included normally. I guess if so just change that inclusion condition from len(entries) == 3 to len(entries) >= 3

An example line is 9.64748 3/31/2025 -82.54 #* -t1 from row 1312921