NucPy

Overview of NucPy

NucPy is a suite of bioinformatics tools for the quantification and analysis of nuclear features from high-content imaging data.

How to download NucPy

# Using git
git clone https://github.com/f-hamidlab/nucpy.git
# or if you have SSH setup
# git clone [email protected]:f-hamidlab/nucpy.git

How to install NucPy

cd nucpy
pip install .

How to run NucPy

Using interactive Jupyter Notebooks

The easiest way to execute NucPy is by using our interactive Jupyter Notebooks for nuclei segmentation and downstream analyses. To use NucPy this way, be sure to load Jupyter on your local computer and open the Notebook you wish to run.

Using command-line terminal

Alternatively, users may run NucPy in Terminal. Below, we provide a quickstart guide to performing nuclei segmentation on our sample image data:

## load modules
import nucpy.segmentation as ncp
import matplotlib.pyplot as plt

## create Segmentador object
path = "../data/sample_images/experiment1"
nps = ncp.Segmentador(path, outdir = None, analyse_all=True)

## set channels and perform nuclear segmentation
nps.set_channels(channels = ["DAPI","Beta3","RFP","Ngn"], marker = "DAPI")
nps.nuclear_segmentation(method = "cellpose", 
                         diameter = 30, 
                         gamma_corr = 0.25, 
                         dc_scaleCorr = 1.9,
                         GPU = True)
                         
## quantify nuclear features
nps.nuclear_features()
nps.add_nuclear_features()
nps.find_dna_peaks(box_size = 10, zoom_box_size = 200)
nps.find_dna_dots(zoom_box_size = 200)
nps.spatial_entropy(d = 5, zoom_box_size = 200)

## output data
nps.saveArrays()
nps.saveChannelInfo()
nps.export_csv(filename = "nucpy_output.csv")