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
Added a new parameter called --ivar_trim_offset to be able to customise this behaviour. As suggested in the issues linked to above if you are using the SWIFT/SNAP protocol then you will need to set --ivar_trim_offset 5 when running the pipeline.
The pipeline will also now attempt to auto-detect whether you are using the SWIFT/SNAP protocol based on the assumption that you have used covid19genome in the name field of the primer BED file. This name is used in the original SWIFT master file containing the primer co-ordinates.
1. Changed clipping penalty parameter in bwa mem from 5 to 1. This favors more soft-trimming at the ends of reads. Since variants at the ends of reads are less reliable, this reduces false-positive variant calling.
2. Expanded ivar trim trimming using the -x offset parameter. This trims extra bases around primer binding sites. This also makes variant calling more stringent.
See these issues for more information on these settings:
nf-core/viralrecon#170andersen-lab/ivar#88
Came across these improvements for ivar in the IDseq github:
chanzuckerberg/idseq-workflows@d922823
andersen-lab/ivar#88
When processing swift libraries polymerases can add additional bases causing issues with trimming. I guess this happens for swift libraries only?
The text was updated successfully, but these errors were encountered: