Gradient Biotech

File formats

Connectivity matrices

CSV

Square n×n numeric matrix, comma-separated. No header row required. Values are edge weights (correlation, partial correlation, or binary adjacency).

Example (5×5 binary):

0,1,1,0,1
1,0,1,1,0
1,1,0,1,0
0,1,1,0,1
1,0,0,1,0

NPZ / NumPy

  • NPZ: array stored under key matrix, conn, or W, or as the sole array in the archive
  • NPY: single 2D array file

MATLAB .mat

Variable named matrix, conn, or W containing a square numeric array.

EEG / MEG

EDF (.edf)

European Data Format multi-channel recording. Channel count and sampling rate detected on import.

BrainVision (.vhdr)

Upload the header file; companion .eeg and .vmrk files should be in the same directory on the server after upload.

EEGLab (.set)

MATLAB structure exported from EEGLab.

fMRI

NIfTI (.nii, .nii.gz)

Preprocessed BOLD timeseries in standard neuroimaging format. Used by functional connectivity and dynamic FC pipelines.

BIDS ZIP

ZIP archive of a BIDS-compliant dataset. Must include dataset_description.json and subject/session folder structure. Run Validate BIDS before FC analysis.

Subject metadata CSV

ColumnRequiredPurpose
external_idYesSubject identifier (S001, …)
age, sex, diagnosis, group_labelNoStandard cohort fields
Any other columnsNoStored as flexible metadata

Task events (BIDS .tsv)

Uploaded in the Task GLM panel on an fMRI dataset.

ColumnRequiredPurpose
onsetYesEvent onset (seconds)
durationYesEvent duration (seconds)
trial_typeYesCondition label

Registry CSV (atlas / montage)

Uploaded on the Atlas & montage registry page (/areas/neurology/registry); validated on upload.

Custom atlas CSV:

ColumnRequiredPurpose
labelYesRegion name (unique)
x, y, zYesMNI centroid (mm)
hemisphere, networkNoRegion grouping metadata

Custom montage CSV:

ColumnRequiredPurpose
nameYesChannel name (unique)
x, yYesSensor position
zNo3D position component

Artifact outputs

Pipeline results are JSON files under data/neuro/artifacts/{run_id}/:

ArtifactContents
metrics.jsonGlobal and per-node graph metrics
connectivity_matrix.jsonProcessed matrix for visualization
eeg_features.jsonBand power and connectivity summaries
nbs_report.jsonNBS significant components
interpretation.jsonAI interpret output

Size guidance

Data typeApprox. size
200×200 connectivity matrix CSV~1 MB
Resting-state fMRI NIfTI (preprocessed)50–500 MB
High-density EEG (1 h @ 256 Hz)50–200 MB

Large uploads are processed on the server; the frontend renders downsampled previews and summary statistics only.