nick-youngblut
commented
4 years ago Even if I provide new reads in the config, the assembly is still skipped if the output directories/files exist, which is a weird behavior:
============================== check_reads ==============================
1 chunks to run. Starting...
2020-07-20 20:23:21 - INFO - index fastq /path/to/files/reads.fq
2020-07-20 20:23:35 - INFO - get seed contigs from input assembly
2020-07-20 20:23:35 - INFO - 8939 total inputs seeds covering 9452078 bases
2020-07-20 20:23:35 - INFO - 89 input seed contigs >= 400bp and >= 10.0x coverage covering 3024691 bases
2020-07-20 20:23:35 - INFO - created 90 bins from seeds
2020-07-20 20:23:35 - INFO - done
--> check_reads completed.
============================== subassemble_reads ==============================
--> 0 chunks need to be run. Skipping...
============================== assemble_olc ==============================
--> 0 chunks need to be run. Skipping...
If the output files/directories exist, athena-meta will just skip those steps. For example:
...but there is no option to force re-run from the start of the workflow. This could be dangerous for anyone thinking that they have re-run athena, but in fact athena just found the old files/directories and skipped everything. It's also a bit annoying to have to manually delete the output files/directories for each re-run.