accord-rs

To bring something into accord (/əˈkɔrd/) is to make it agree or correspond. The accord-rs library calculates a consensus from reads that where aligned against a reference sequence.

This library is in the early alpha stage, and its API may be subject to change without notice.

Python

Accord provides bindings for consensus calculation from Python code. Install with pip install accord-rs, and then use like so:

from accord import Calculator
from accord.data import AlnQualityReqs

# settings for consensus calculation
reqs = AlnQualityReqs(
    min_mapq=10,            # minimum mapping quality
    mandatory_flags=0,      # required SAM flags, see: https://broadinstitute.github.io/picard/explain-flags.html
    prohibited_flags=1540,  # no unmapped, failing quality or optical duplicate reads
    indel_cutoff=0.1,       # only indels contained in at least 10 % of reads covering that position are considered
    save_ends=0,            # this has no function yet - only relevant when using PCR fragments
    min_observations=50,    # base coverage needs to be at least 50
)

# generate the consensus and alignment statistics
calc = Calculator("/path/to/reference.fasta", "/path/to/aln.bam", reqs)
consensus, stats = calc.calculate()

# print the consensus as a FASTA record
print(consensus.to_fasta())