scprel 1.2

Single-cell data preprocessing for multiple samples.

Scprel - Single-Cell Data Preprocessing in Python

Import scprel as:

import scprel

This package allows to perform basic preprocessing steps for single-cell analysis of multiple samples. It includes scrublets detection, quality control, normalization with target_sum=1e4, leiden clustering, annotation of cell types with PanglaoDB database and infercnv calculations. It integrates some of the Scanpy, Decoupler, Infercnvpy and Anndata functions. It is designed to facilitate workflow when analyzing multiple samples. Each sample is analyzed and annotated separately and then all samples are concatenated.

Example of usage:

scprel.scrun(names = ['sample1', 'sample2'], path = '/content/drive/MyDrive/MyDirectory/')

• names - list of sample names in your directory (.h5 format);
• path - path to the directory with samples

The result of this function is the anndata.AnnData object, compressed with hdf5plugin, with concatenated samples, filtered by 'mt' and 'ribo' genes, with annotated gene locations and annotated tumor cells based on cnv score. All immune cells in the sample are considered reference cells for infercnv calculations.

The resulting file will be saved in your default home directory and is ready for batch correction and further analysis.