Tutorial 3: Merging Timeseries With neurocaps.analysis.merge_dicts

Combining the timeseries from different tasks is possible with merge_dicts, this permits running analyses to identify similar CAPs across different tasks, assuming these tasks use the same subjects. The merge_dicts() function will produce a merged subject timeseries dictionary that contains only the subject IDs present across both subject dictionaries. Additionally, this function appends similar run-IDs together. For instance, run-1 from one task is appended to run-1 of the other task. For this to work, all dictionaries must contain the same number of columns/ROIs.

import numpy as np
from neurocaps.analysis import merge_dicts

# Using a random dictionary that would be generate by TimeseriesExtractor to demonstrate what merge_dicts does
# Generating a subject_timeseries dict containing five subjects, each with three runs that have 10 timepoints
# and 100 rois.
subject_timeseries_1 = {str(x) : {f"run-{y}": np.random.rand(10, 100) for y in range(1, 4)} for x in range(1, 6)}
# Deleting run-2 for subject 2; situation where subject 2 only completed two runs of a task
del subject_timeseries_1["2"]["run-3"]
# Second dictionary contains 2 subjects, with a single run, each with 20 timepoints
subject_timeseries_2 = {str(x) : {f"run-{y}": np.random.rand(20, 100) for y in range(1, 2)} for x in range(1, 3)}

# subject_timeseries_list also takes pickle files and can save the modified dictionaries as pickles too.
subject_timeseries_merged = merge_dicts(subject_timeseries_list=[subject_timeseries_1,
                                                                 subject_timeseries_2],
                                        return_merged_dict=True,
                                        return_reduced_dicts=False)

for subj_id in subject_timeseries_merged["merged"]:
    for run_id in subject_timeseries_merged["merged"][subj_id]:
        timeseries = subject_timeseries_merged["merged"][subj_id][run_id]
        print(f"sub-{subj_id}; {run_id} shape is {timeseries.shape}")

sub-1; run-1 shape is (30, 100)
sub-1; run-2 shape is (10, 100)
sub-1; run-3 shape is (10, 100)
sub-2; run-1 shape is (30, 100)
sub-2; run-2 shape is (10, 100)

# The original dictionaries can also be returned too. The only modifications done is that the originals will
# Only contain the subjects present across all dictionaries in the list
merged_dicts = merge_dicts(subject_timeseries_list=[subject_timeseries_1,
                                                    subject_timeseries_2],
                           return_merged_dict=True,
                           return_reduced_dicts=True)

for dict_id in merged_dicts:
    for subj_id in merged_dicts[dict_id]:
        for run_id in merged_dicts[dict_id][subj_id]:
            timeseries = merged_dicts[dict_id][subj_id][run_id]
            print(f"For {dict_id} sub-{subj_id}; {run_id} shape is {timeseries.shape}")

For dict_0 sub-1; run-1 shape is (10, 100)
For dict_0 sub-1; run-2 shape is (10, 100)
For dict_0 sub-1; run-3 shape is (10, 100)
For dict_0 sub-2; run-1 shape is (10, 100)
For dict_0 sub-2; run-2 shape is (10, 100)
For dict_1 sub-1; run-1 shape is (20, 100)
For dict_1 sub-2; run-1 shape is (20, 100)
For merged sub-1; run-1 shape is (30, 100)
For merged sub-1; run-2 shape is (10, 100)
For merged sub-1; run-3 shape is (10, 100)
For merged sub-2; run-1 shape is (30, 100)
For merged sub-2; run-2 shape is (10, 100)