Bam2cell

A package to split a BAM file based on cell type annotation in AnnData objects.

Usage and Examples

There are two modes sequential and parallel. The sequential mode will process cell types one by one but is more disk space friendly, the parallel is more disk space hungry but faster, since it process all cell types at the same time.

A minimal example is shown here:

⚠️ Note: The barcodes should not contain suffix or prefix. Use clean_bcs() to remove them.

import bam2cell
import anndata as ad

adata = ad.read_h5ad("data/adata.h5ad")

generator = bam2cell.GenerateCellTypeBAM(adata, 
                                         annot_key="annotation",
                                         output_path="data/",
                                         input_bam="data/AllCellsSorted_toy.bam",
                                         tmp_path="data/",
                                         workers=8,
                                         )
generator.process_all_parallel()  # Case 1 - Process all cell types at the same time
generator.process_cts_sequential() # Case 2 - Process cell types one by one

For a more advanced usage, you can use the function bam2cell, which allow to process an AnnData with multiple samples.

import bam2cell
import anndata as ad
import pandas as pd

adata = ad.read_h5ad("data/adata.h5ad")
artificial_batch = ["batch1"] * 100 + ["batch2"] * 91
adata.obs["batch"] = pd.Categorical(artificial_batch)
adata.obs["bam_path"] = "data/AllCellsSorted_toy.bam"

bam2cell.bam2cell(adata,
                  annot_key="annotation",
                  input_bam=None,  # Only when we have 1 batch in the AnnData
                  output_path="data/",  
                  tmp_path="data/",
                  bam_key="bam_path",  # For each barcode we have the path to the BAM file
                  batch_key="batch",  
                  mode="parallel",
                  suffix=None,  # Suffix in the barcode to be removed (e.g., BC-1-suffix --> BC-1)
                  prefix=None,  # Prefix in the BC to be removed (e.g., prefix-BC-1 --> BC-1) 
                  workers=8
                  )

Installation

You need to have Python 3.10 or newer installed on your system. There are several alternative options to install bam2cell:

1. Install the latest release of bam2cell from PyPI:

pip install bam2cell  

2. Install the latest development version:

pip install git+https://github.com/davidrm-bio/bam2cell.git@main