refine-results
It refines the read-by-read classification file produced by classify. It supports filtering by score, excluding or selecting specific taxa, adjusting rank, and more.
Usage
metabuli refine-results <i:classifications> <i:DBDIR> [options]
| Argument |
Description |
classifications |
The JOB_ID_classifications.tsv file generated by the classify step |
DBDIR |
The same database directory used in the classify step |
Options
| Option |
Description |
--min-score |
Remove classifications with a score below this value |
-e |
Remove classifications with large e-value (0: disable) |
--remove-unclassified |
Remove unclassified reads from the output |
--exclude-taxid |
Remove the specified taxa as well as its children (e.g., 1758,9685) |
--select-taxid |
Remain only the specified taxa as well as children (e.g., 1758,9685) |
--select-columns |
Columns to print (0-based csv) |
--report |
Write a report of the refined classification results |
--rank |
Adjust all classifications to the specified rank |
--rank-file-type |
How to handle classifications at ranks higher than --rank. 0 = exclude, 1 = include, 2 = write to a separate file |
--threads |
Number of threads to use (all by default) |
Output Files
| File |
Description |
JOB_ID_refined.tsv |
Refined per-read classification results |
JOB_ID_refined_report.tsv |
Report of refined results (with --report) |
JOB_ID_refined_krona.html |
Krona chart of refined results (with --report) |
JOB_ID_refined_higherRanks.tsv |
Classifications at ranks above --rank (with --rank-file-type 2) |