Data Types¶
Synapse uses a set of typed data containers that flow between nodes. Each type has a distinct port color.
ImageData (Green)¶
A 2D image represented as a PIL Image.
- Formats: Grayscale (L), RGB, RGBA
- Bit depth: 8-bit (uint8) or 16-bit (uint16)
- File types: TIFF, PNG, JPEG, BMP, and more
- Conversion: Use "RGB to Gray" to convert color to grayscale
MaskData (Forest Green)¶
A binary mask — a single-channel image where white (255) represents foreground and black (0) represents background.
- Created by threshold nodes, ROI nodes, or segmentation tools
- Used as input for morphological operations and measurements
- Multiple masks can be combined with Image Math
LabelData (Chartreuse)¶
An integer-valued image where each connected region has a unique label (1, 2, 3, ...).
- Created by Watershed, Label 2D, or segmentation nodes
- Background pixels have label 0
- Displayed as a color-coded overlay
- Used for per-object measurements
SkeletonData (Yellow-Green)¶
A thinned skeleton mask — a special subtype of MaskData produced by skeletonization.
- Created by Skeletonize node
- Used for branch analysis, filopodia measurement, and path tracing
TableData (Blue)¶
A tabular dataset (pandas DataFrame) with named columns and rows.
- Created by measurement nodes (Particle Props, Image Stats, etc.)
- Manipulated with Data & Table nodes (Filter, Sort, Aggregate, etc.)
- Visualized with Plot nodes
- Saved to CSV/Excel with the Data Saver node
StatData (Royal Blue)¶
A statistical result object containing test statistics, p-values, and summary tables.
- Created by analysis nodes (Grouped Comparison, Pairwise Comparison, etc.)
- Displayed in the node's output or connected to a Data Table Node
FigureData (Purple)¶
A matplotlib figure object for scientific plots.
- Created by Plot nodes (Scatter, Histogram, Box Plot, etc.)
- Edited with the Figure Editor node
- Displayed inline with Data Figure Node
- Exported as SVG/PNG/PDF via the SVG Editor node
PathData (Gray)¶
A file or folder path string.
- Produced by Folder Iterator (one path per batch iteration)
- Consumed by Image Reader, Table Reader, etc.
- Modified with Path Modifier node