Tag Extraction #
Finding ’things’ in reads, and then using that information is called ’tagging'.
Extraction and downsteam steps are tied together with arbitrary ’labels’, which are split into in(put) and out(put) labels.
This allows you to efficiently perform multiple actions with one search, for example trim adapter tags and keep only reads that contain the adapter.
mbf-fastq-processor errors early if a step introduces a label that is never used or removed by later transformations.
Tags generating steps are split into three sections, depending on their output:
- calc for steps that create numeric labels,
- convert for steps that reshape existing tags into new ones,
- extract for steps that define ‘regions’ within your reads,
- tag section for steps that add boolean values to your reads.
Finally, see and the tag usage section for helpers that consume or export those labels.
There are further tag using steps in the modification steps and filter steps sections.
Rules for Tag labels #
Tag labels must conform to ‘[a-zA-Z_][a-zA-Z0-9_]*$’ and are case sensitive (this is so they can be used in EvalExpression.
Tag labels may not be ‘ReadName’ (first column in StoreTagsInTable), nor may they start with ’len_’ (used by EvalExpression as virtual tags.