You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
The pipeline now queries the NCBI Taxonomy API rather than GoaT to establish the list of lineages on which to run BUSCO. The possible lineages are now defined in the pipeline configuration but can be overridden with the --lineage_tax_ids parameter.
The pipeline will now immediately fail if the assembly can't be retrieve by the datasets command-line tool.
Pipeline information is now outputted in pipeline_info/genomenote/ instead of genomenote_info/.
maxRetries increased to 5 to cope with large datasets.
BUSCO now runs in "scratch" mode, i.e. off a temporary directory, as the number of files it creates could otherwise overwhelm a network filesystem.
SORT, FASTK, and MERQURYFK, can now put their temporary files in the work directory rather than /tmp. Turn that on with the --use_work_dir_as_temp flag.
The memory requirement of SORT is adjusted to account for some overheads and avoid the job to be killed.
All resource requirements (memory, time, CPUs) now fit the actual usage. This is achieved by automatically adjusting to the size of the input whenever possible.
Genomes with sequences longer than 2 Gbp are now supported thanks to upgrading FastK and MerquryFK.
Fixed a bug that was causing the Completeness to be reported as 0 in the statistics CSV file, when the k-mer database was constructed from BAM files.
Minimal version of Nextflow downgraded to 23.04 to 22.10. 22.10 is tested as part of our continuous integration (CI) pipeline.
The "test" profile now runs faster, thanks to tuning some Busco/Metaeuk parameters.
The "test_full" profile is now tested automatically when updating the dev and main branches.