Skip to content

Utility

Editable Table

Displays an input table in an editable spreadsheet widget and outputs the modified result.

Details

The node accepts a TableData input and presents it in an Excel-like spreadsheet where you can type anywhere to add data, double-click column headers to rename them, single-click headers to sort, and right-click for insert/delete operations. Copy/paste and Delete key are supported. Changes are pushed downstream automatically.

Parameters:

  • Reset Edits on Next Run — when checked, discards local edits and reloads from upstream on the next evaluation
Direction Port Type
Input in table
Output out table

Properties: Reset Edits on Next Run

Extract Object

Extracts a single object (image, figure, label, etc.) from a table's object column.

Details

After batch-accumulating images or figures, this node lets you pick one item by row index and outputs it as its original data type.

Parameters:

  • Row Index — 1-based row number to extract from
  • Object Column — name of the column containing the objects (default: object)
Direction Port Type
Input in table
Output out any

Properties: Row Index (1-based)

Blank Subtract

Subtract a reference row's value from all rows in a column.

Details

Common use: subtract background (BG) measurement from all cells.

  • Reference Column — the column containing group/cell labels (e.g. 'cell')
  • Reference Value — the label of the reference row (e.g. 'BG')
  • Target Columns — columns to subtract from (comma-separated, or leave empty for all numeric)

The reference row's value is subtracted from every row in each target column. The reference row itself is kept (will become 0).

Direction Port Type
Input in table
Output table table

Row Normalize

Normalize all rows by a reference row's value (divide instead of subtract).

Details

Useful for fold-change calculations: value / reference.

  • Reference Column — column containing group labels
  • Reference Value — label of the reference row (e.g. 'Control', 'BG')
  • Target Columns — columns to normalize (comma-separated, or empty for all numeric)
Direction Port Type
Input in table
Output table table

Drop Rows

Drop rows where a column matches any of the specified values.

Details
  • Column — which column to check
  • Values to Drop — comma-separated list of values to remove (e.g. 'BG, Artifact (BG)')
Direction Port Type
Input in table
Output table table

Python Script

Run custom Python code with dynamic input and output ports.

Details

Use this node for operations that no dedicated node covers — custom formulas, advanced scipy/skimage functions, string parsing, conditional logic, or any one-off data transformation.

Setup

  • Inputs / Outputs spinboxes control how many ports the node has.
  • Click Edit Script… to open the full code editor (dark theme).
  • The inline preview on the node card shows the current script.
  • print() output is shown as a popup after execution.

Variables

Variable Description
in_1, in_2, … Data from each input port (DataFrame, ndarray, or raw value). Unconnected = None.
out_1, out_2, … Assign results here to send downstream.
pd pandas
np numpy
scipy scipy (use scipy.stats, scipy.ndimage, etc.)
skimage scikit-image (use skimage.filters, etc.)
cv2 OpenCV
PIL Pillow
plt matplotlib.pyplot
set_progress(0-100) Update the node's progress bar during long operations

You can import any additional module installed in your environment.

Output types

Results are auto-wrapped: DataFrame → TableData, 2D ndarray → ImageData, Figure → FigureData, scalar → single-cell TableData. To force a type, use: out_1 = MaskData(payload=arr) or ImageData(payload=arr, bit_depth=16).

Examples

  • Fold-change (qPCR) — df['fold_change'] = 2 ** (-df['ddCt']):

  • df = in_1.copy()

  • df['fold_change'] = 2 ** (-df['ddCt'])

  • out_1 = df

  • Column ratio — df['ratio'] = df['intensity'] / df['area']:

  • df = in_1.copy()

  • df['ratio'] = df['intensity'] / df['area']
  • out_1 = df

Split by median (set Outputs to 2):

  • med = in_1['value'].median()
  • out_1 = in_1[in_1['value'] > med]
  • out_2 = in_1[in_1['value'] <= med]

Custom scipy test:

  • from scipy.stats import mannwhitneyu
  • g1 = in_1[in_1['group']=='A']['value']
  • u, p = mannwhitneyu(g1, g2)
  • out_1 = pd.DataFrame({'U': [u], 'p': [p]})

Image filter:

  • from scipy.ndimage import gaussian_filter
  • out_1 = gaussian_filter(in_1, sigma=3)
Direction Port Type
Input in_1 any
Output out_1 any

Properties: Inputs, Outputs