Usage

To diplay options, simply run the following line in terminal:

$ branesta -h

Usage:

branesta [-h] [--subnetpath SUBNETPATH] [--tot_len TOT_LEN]
         [--tot_roi_num TOT_ROI_NUM]
         srcdir outdir win_len

positional arguments:
  srcdir                path to source directory with time-series
  outdir                path to out directory for results and logs
  win_len               number of time-frames in each snapshot

optional arguments:
  -h, --help            show this help message and exit
  --subnetpath, -s
                        path to csv file with subnetwork labels and ixs,
                        optional
  --tot_len, -l
                        expected total number of time frames, optional
  --tot_roi_num, -r
                        expected total number of ROIs, optional

Options are explained below in more details:

  • srcdir: path to times-series to analyze. There are certain requirements for the input time-series:

  1. Must be parcelled into ROIs, should not be more than a few hundred ROIs.

  2. Shape must be 2D: time x space

  3. File format must be .csv with no header row or index column

  4. Filenames must follow BIDS convention (https://bids.neuroimaging.io/) (see example time-series for reference)

  • outdir: path to output directory. Computed metrics and log files will be placed here.

  • win_len: number of time-frames within each time window. With a TR of ~1s, at least 30 is recommended. The more regions we use, the more time-frames we need to properly estimate the covariance matrix

  • –subnetpath: optional, path to subnetwork indexes file. If provided, network stability is computed separately for edges corresponding to subnetworks. If not provided, the program defaults to a default subnetwork (labeled as whole_brain) consisting of all ROIs. Must be a .csv with two columns and a header row (roi, network). (see example subnetworks_willard.csv for reference)

  • –tot_len: optional, expected length of input time-series. If provided together with tot_roi_num, the program checks for proper shape of every input file.

  • –tot_roi_num: optional, expected number of spatial ROIs. If provided together with tot_len, the program checks for proper shape of every input file.

Example:

$ branesta data/derivatives/parcelled data/derivatives/network_stability 30 -s utils/subnetworks.csv -l 720 -r 498

To use branesta in a python shell:

import branesta.branesta as bns

bns.analyze?

We plan to add more details here. Contact Botond Antal if you need help.