Creating a manifest from another tabular file for a cross-sectional study

In this example, we have a cross-sectional study where not all participants have the same MRI modalities (datatypes).

The CSV file example1-participants.csv (shown below) indicates whether participants have diffusion or functional data. All participants have anatomical data.

participant	age	sex	dwi	func
001	25	F	True	True
002	26	F	False	True
003	27	F	True	False
004	28	F	False	False
005	29	M	True	True
006	30	M	False	True
007	31	M	True	True
008	32	M	True	False

Here is a script that creates a Nipoppy manifest from the above file:

Attention

The script below was written for Python 3.11 with pandas 2.2.3. It may not work with older/different versions.

#!/usr/bin/env python
"""Manifest-generation script for Example 1."""

from pathlib import Path

import pandas as pd

if __name__ == "__main__":
    # get the path to the participants file
    # we assume that it is in the same directory as this script
    path_participants = Path(__file__).parent / "example1-participants.csv"

    # load the participants file
    # note that the participant column is read as a string because
    # otherwise the leading zeros would be removed
    df_participants = pd.read_csv(path_participants, dtype={"participant": str})

    data_for_manifest = []
    for _, row in df_participants.iterrows():
        # no change for participant_id
        participant_id = row["participant"]

        # use the same visit_id and session_id for all participants
        # since the study is cross-sectional
        visit_id = 1
        session_id = 1

        # all participants have anat data
        datatype = ["anat"]
        if row["dwi"]:
            datatype.append("dwi")
        if row["func"]:
            datatype.append("func")

        # create the manifest entry
        data_for_manifest.append(
            {
                "participant_id": participant_id,
                "visit_id": visit_id,
                "session_id": session_id,
                "datatype": datatype,
            }
        )

    df_manifest = pd.DataFrame(data_for_manifest)

    # write the manifest in the same directory as this script
    df_manifest.to_csv(
        Path(__file__).parent / "example1-manifest.tsv", sep="\t", index=False
    )

Running this script creates a manifest that looks like this:

participant_id	visit_id	session_id	datatype
001	1	1	[‘anat’, ‘dwi’, ‘func’]
002	1	1	[‘anat’, ‘func’]
003	1	1	[‘anat’, ‘dwi’]
004	1	1	[‘anat’]
005	1	1	[‘anat’, ‘dwi’, ‘func’]
006	1	1	[‘anat’, ‘func’]
007	1	1	[‘anat’, ‘dwi’, ‘func’]
008	1	1	[‘anat’, ‘dwi’]