PRISM analysis

• program to process and analyze PRISM images of neuronal synapses

This program takes the PRISM images acquired using Perkin Elmer Opera Phenix High-Content Screening System as input,

• performs drift and flatfield corrections
• parse images from the same well and save as individual MAT files
• segment the images and save the masks
• extract features of synapses
• perform t-SNE and hierarchical clustering analysis on extracted synaptic features

Example images will be available soon

Usage

• Process PRISM images

  Unzip the files and add the folder locaton to MATLAB search path. Change params.parentFolderForAnalysis in main.m to the location of the folders with raw images, and run main.m. Processed images will be saved at the path specified by params.outputImgsPath.

• Analysis

  Run synapse_t-SNE.py for t-SNE analysis and synapse_clustering.py for hierarchical clustering. Change `dir_path' in the script to the folder where the MAT files are located, 'fig_path' to the desired figure output path.

Output Data structure

Each Rep# folder contains one biological replicate (neuronal culture, plate, or batch), total 3 batches. Each sub-folder RepX-Y contains one replicate (well) within each biological replicate, total 3 wells.

• MultiplexImageData.mat: parsed, unprocessed images

  imgsProj: collection of aligned, top-hat filtered images from a single well, has the structure imgsProjTopHat{field of view}{channel, imaging round}, channel order:'MAP2','DAPI','LNA','vGlut' imaging round: see variable 'imgRoundNames'

  imgRoundNames: list of names of incubated probe sequences (p#), wash-out (WO) or not, exposure time (#S).

  List of correspoding protein targets: {'actin','p2';'Tuj-1','p3';'cortactin','p4';... 'Shank3','p6';'ARPC2','p7'; 'bassoon','p8'; 'synapsin-1(2)','p9-t2'; 'synapsin-1','p9';... 'Homer-1b/c','p10';'NR2B','p12';'PSD-95','p1-'} ;

• MultiplexImageDataAligned.mat: processed images after drift correction and flatfiel correction.

• MultiplexImageAlignedTopHat.mat: same as MultiplexImageDataAligned.mat but with top-hat filtering.