A package to analyze celltypes on high definition spatial profiling assays

Installation

It is recommended to install the package in a virtual environment or a Conda environment. To create a Conda environment, run the following command:

conda create -n easydecon python=3.10.14
conda activate easydecon

You can install from PyPi:

pip install easydecon

To install directly from GitHub using pip into the active environment, run the following command:

pip install git+https://github.com/sinanugur/easydecon.git

Overview

Absolute Minimal Example

from easydecon.easydecon import *
from easydecon.config import *
from easydecon.extra import *

#read your DESeq table into a markers_df
#sdata is your VisiumHD file in SpatialData format or segmented AnnData object, assumed you QC and etc.
markers_df=read_markers_dataframe(sdata,filename="scanpy_deseq_table.csv")

#run easydecon
ph1, ph2, assigned_labels, posterior_df, proportions_df= easydecon_workflow(sdata,markers_df=markers_df)

#or setting prior genes
ph1, ph2, assigned_labels, posterior_df, proportions_df= easydecon_workflow(sdata,markers_df=markers_df,marker_genes=["gene1","gene2","gene3"])

`assigned_labels` will be added to sdata.obs and contain the celltype assignments
`proportions_df` will contain the estimated proportions for each celltype

Usage and Documentation

You may find our example notebooks in the notebooks folder.

• Demo notebook for a single-cell Anndata object (demo)[https://github.com/sinanugur/easydecon/blob/main/notebooks/demo.ipynb]
• Demo notebook for macrophage markers (demo_macrophage)[https://github.com/sinanugur/easydecon/blob/main/notebooks/demo_macrophage.ipynb]
• Minimal example notebook (minimal)[https://github.com/sinanugur/easydecon/blob/main/notebooks/demo_minimal_example.ipynb]

Segmentation Example

• Segmentation with bin2cell example [https://github.com/sinanugur/easydecon/blob/main/notebooks/demo_bin2cell_minimal.ipynb]