USAGE

Pipeline for TractoInferno evaluation
=====================================

USAGE

nextflow run main.nf -resume --input ... --reference ... --atlas ... -with-docker scilus/scilus:1.2.0_rbxflow-1.1.0 [OPTIONAL_ARGUMENTS]

The atlas used for TractoInferno is available at https://zenodo.org/record/4630660#.YJvmwXVKhdU

DESCRIPTION

    --input=/path/to/[root]             Root folder containing multiple subjects and tractograms (.tck or .trk).

                                        [root]
                                        ├── S1
                                        │   └── *.trk
                                        └── S2
                                            └── *.trk

    --reference=/path/to/[root]         Root folder containing multiple subjects and candidate tractograms

                                        [root]
                                        ├── S1
                                        │   ├── tractography
                                        │   │   ├── S1__AF_L.trk
                                        │   │   ├── S1__AF_R.trk
                                        │   │   ├── S1__CC_FR_1.trk
                                        │   │   └── ...
                                        │   └── dti
                                        │       └── S1__fa.nii.gz
                                        └── S2
                                            ├── tractography
                                            │   ├── S2__AF_L.trk
                                            │   ├── S2__AF_R.trk
                                            │   ├── S2__CC_FR_1.trk
                                            │   └── ...
                                            └── dti
                                                └── S2__fa.nii.gz

    --atlas                             Path to TractoInferno RBX atlas

    OPTIONAL ARGUMENTS (current value)

    --register_processes                    Number of processes for registration task ($register_processes).
    --rbx_processes                         Number of processes for bundle recognition task ($rbx_processes).
    --processes                             The number of parallel processes to launch ($cpu_count).
                                            Only affects the local scheduler.

    --output_dir                            Directory where to write the final results.
                                            By default, will be in "./results_rbx"


AVAILABLE PROFILES (using -profile option (e.g. -profile large_dataset,cbrain))

large_dataset                               When this profile is used, rbx_flow will be asking for 16 GB instead of 8 GB.

fully_reproducible                          When this profile is used, all the parameters will be set to have 100% reproducible results.

macos                                       When this profile is used, rbx_flow will modify a parameter (scratch) for MacOS users.

cbrain                                      When this profile is used, Nextflow will copy all the output files in publishDir and not use symlinks.

NOTES

Use '-C config_file.config' to specify a non-default configuration file.
The '-C config_file.config' must be inserted after the nextflow call
like 'nextflow -C config_file.config run ...'.