bulkrna-read-qc

bulkrna-read-qc

When to use

Read raw Phred+33 FASTQ before alignment. This implements selected quality metrics, not the complete FastQC suite. Use bulkrna-read-alignment for logs.

Use from a step

from skills._sdk.notebook import load_skill, write_output
library = load_skill('bulkrna-read-qc')
data = library.demo_data(random_state=42)
result = library.quality_control(data)
write_output(result, 'tables/qc_summary.csv')

For real files, pass read_fastq, read_log or read_reference as appropriate to read_input(..., reader=...). examples/example_step.py is executable.

API

read_fastq(path: str | Path, *, max_reads: int=100000) -> pd.DataFrame

Read FASTQ records; pass this function as reader= to read_input.

:param path: FASTQ path; a .gz suffix selects gzip decompression. :param max_reads: CLI limit 100000; increase for a larger leading-read sample. :returns: Sequence and Phred+33 quality strings as rows. :raises ValueError: Records are truncated, malformed, or have unequal sequence/quality lengths.

quality_control(reads: pd.DataFrame, *, max_reads: int=100000) -> pd.DataFrame

Compute a new one-row quality summary for Phred+33 reads.

:param reads: sequence and quality string columns; input rows are not changed. :param max_reads: CLI default 100000 leading records; increase to inspect more reads. :returns: Scalar QC metrics and adapter counts; run_info retains per-base distributions. :raises ValueError: Reads, lengths or Phred+33 scores are invalid.

run_info(result: pd.DataFrame, *, keep: bool=True) -> dict

Read full quality distributions and sampling diagnostics.

:param result: Output of quality_control. :param keep: True preserves attrs; False removes diagnostics before serialization. :returns: A separate dictionary containing metrics and sampling provenance. :raises TypeError: The result is not a DataFrame.

quality_figure(result: pd.DataFrame)