dnaio is a Python 3.8+ library for very efficient parsing and writing of FASTQ and also FASTA files.
Since dnaio version 1.1.0, support for efficiently parsing uBAM files has been implemented.
This allows reading ONT files from the dorado
basecaller directly.
The code was previously part of the Cutadapt tool and has been improved significantly since it has been split out.
The main interface is the dnaio.open function:
import dnaio
with dnaio.open("reads.fastq.gz") as f:
bp = 0
for record in f:
bp += len(record)
print(f"The input file contains {bp/1E6:.1f} Mbp")
For more, see the tutorial and API documentation.
Using pip:
pip install dnaio zstandard
zstandard can be omitted if support for Zstandard (.zst) files is not required.
- FASTQ input and output
- FASTA input and output
- BAM input
- Compressed input and output (
.gz,.bz2,.xzand.zstare detected automatically) - Paired-end data in two files
- Interleaved paired-end data in a single file
- Files with DOS/Windows linebreaks can be read
- FASTQ files with a second header line (after the
+) are supported
- Multi-line FASTQ files are not supported
- FASTQ and uBAM parsing is the focus of this library. The FASTA parser is not as optimized