Operations

Read a folder of grayscale or RGB .jp2 slice files into split-channel tif series and RGB composites.

Make existing tiff series compatible with the analysis tools

Read multiscale .omehans file format (OME Zarr variation) into a tiff series

Read a piece of .ims Imaris file into a tiff series.

Read multiscale .ims Imaris file format into a tiff series.

Read multiscale .ome.zarr data into a tiff series using the top-level scale arrays.

Apply folder-wise image math or logic to matching TIFF pairs

Denoise image using available image restoration models from cellpose

Remove stripes (mainly for RSCM images)

Apply gamma correction to adjust image intensity

Apply Gaussian blur to smooth the image

Rescale image intensities between 2 and 98%

Apply adaptive histogram equalization to improve local contrast

Resize each TIFF image by a uniform x/y scale factor

Remove image background with SAM2 (one GPU job per z-layer); saves the background-removed image and the brain mask

Remove image background using a neural net

Cluster puncta coordinates based on their distance and density

Compute the distance to the nearest neighboring point in microns (3D) for each point; transform_points-style CSVs also get a per-region mean/median summary

Map detected cells to corresponding atlas parcellations based on existing registration

Compute mean intensity in a sphere around each detected cell position using the raw image resolved from provenance

Delete detected points (noise) that are in the background. Requires path to detected spots (detected with deepblink / cellfinder / other) and path to background/foreground masks (segmented with ilastik / other)

Classify puncta to true cells or artifacts using a binary ResNet-50 deep learning classifier

Semantic segmentation / pixel classification of an arbitrary 3D image stack based on a random forest classifier model

Semantic segmentation of a 3D image stack using a provided model

Detect cells in an arbitrary 3D image volume

Detect puncta in 3D image stack using deep neural network

Detect cells (puncta) in a 3D image stack using threshold-based method

Detect spots (puncta / cells / nuclei) in a 3D image stack using deep learning method

Register brain to an atlas

Register stack representing a 3D brain volume to an atlas from brainglobe atlas API

Add columns describing your data (such as treatment, sex, time point) into the cells csv