Human Workflow

The host-specific branch performs sequential classification of host reads against: - mature miRNA - tRNA - rRNA - other ncRNA

The native pipeline now mirrors the updated legacy human_final_pipeline.sh behavior more closely: - intermediate remaining-read FASTQs are retained in the human subtree - count files use the .human.*.counts.tsv naming pattern - summary generation is delegated to summarize_human_subtree.py

Expected summary output: - <sample>.human.summary.tsv