Home

How to analyze dnbc4tools output in R and Python

scRNA Analysis • scATAC Analysis

R (Seurat)

Load the filtered gene expression matrix using the Read10X function.

library(Seurat)
counts.data <- Read10X(data.dir = "/outs/filter_matrix")

Python (Scanpy)

You can load the data using either the h5ad file or the MEX matrix directory.

1. Read h5ad format (recommended):

import scanpy as sc
adata = sc.read_h5ad('/outs/filter_feature.h5ad')

2. Read matrix format:

import scanpy as sc
adata = sc.read_10x_mtx('/outs/filter_matrix')

R (Signac)

Load the filtered peak matrix and create a Seurat object.

require(magrittr)
require(readr)
require(Matrix)
require(tidyr)
require(dplyr)
library(Signac)
library(Seurat)

# Function to read dnbc4tools ATAC output
read_signac_C4 <- function(mex_dir_path, fragments, singlecellmetadata){
    mtx_path <- paste(mex_dir_path, "matrix.mtx.gz", sep = '/')
    feature_path <- paste(mex_dir_path, "peaks.bed.gz", sep = '/')
    barcode_path <- paste(mex_dir_path, "barcodes.tsv.gz", sep = '/')
    
    features <- readr::read_tsv(feature_path, col_names = F) %>% tidyr::unite(feature)
    barcodes <- readr::read_tsv(barcode_path, col_names = F) %>% tidyr::unite(barcode)
    
    mtx <- Matrix::readMM(mtx_path) %>%
    magrittr::set_rownames(features$feature) %>%
    magrittr::set_colnames(barcodes$barcode)

    metadata <- read.csv(
        file = singlecellmetadata,
        header = TRUE,
        row.names = 1
    )
    
    chrom_assay <- CreateChromatinAssay(
        counts = mtx,
        sep = c("_", "_"),
        fragments = fragments,
        min.cells = 10,
        min.features = 200
    )

    scATAC <- CreateSeuratObject(
        counts = chrom_assay,
        assay = "peaks",
        meta.data = metadata
    )
    return(scATAC)
}

R (ArchR)

Create Arrow files from the fragments.tsv.gz file.

library(ArchR)
ArrowFiles <- createArrowFiles(
  inputFiles = "/outs/fragments.tsv.gz",
  sampleNames = "MySample",
  filterTSS = 4, 
  filterFrags = 1000, 
  addTileMat = TRUE,
  addGeneScoreMat = TRUE
)

Python (AnnData)

Load the filtered peak matrix into an AnnData object.

import os
from scipy.sparse import csr_matrix
import anndata
import pandas as pd
import scipy.io
import scipy.sparse

def read_atac_C4(path):
    mtx_path = os.path.join(path, "matrix.mtx.gz")
    barcode_path = os.path.join(path, "barcodes.tsv.gz")
    feature_path = os.path.join(path, "peaks.bed.gz")

    obs = pd.read_csv(barcode_path, header=None, index_col=0, sep="\t")
    obs.index.name = "barcode"

    var = pd.read_csv(feature_path, header=None, sep="\t")
    var.columns = ['chrom', 'chromStart', 'chromEnd']
    var.index = var['chrom'].astype(str) + ':' + var['chromStart'].astype(str) + '-' + var['chromEnd'].astype(str)
    var.index.name = "peak"

    mtx = csr_matrix(scipy.io.mmread(mtx_path).T)

    adata = anndata.AnnData(mtx, obs=obs, var=var)
    adata.var_names_make_unique()
    return adata