This is a package for reading BCF files (binary call format). Similar to pysam/cyvcf2, but limited to reading from BCF files only, and oriented around getting genotype data into numpy arrays quickly.

'''py from pybcf import BcfReader

bcf = BcfReader(bcf_path)

for var in bcf: # the usual attributes are available e.g. # var.chrom, var.pos, var.ref, var.alts, var.info['AF']

# get genotypes as numpy array
genotypes = var.samples['GT']  # as numpy array, missing=-1

'''

• can't fetch variants from random regions yet
• not extensively tested yet
• extracting info fields is slow at the moment