Closed 24natasya closed 4 years ago
The argument is a (CSV) file path where Octopus will write the data profile to, e.g.:
$ octopus -R ref.fa -I reads.bam -o calls.vcf --data-profile reads.profile.csv
You then use one or more profiles to generate new error models:
$ profiler.py -P reads.profile.csv -O reads.model
Then you can use this error model for variant calling:
$ octopus -R ref.fa -I reads.bam -o calls.vcf --sequence-error-model reads.model
There is no truth set needs to fit the error model.
Thanks for your prompt reply. I really appreciate!
On Wed, Oct 9, 2019 at 4:53 PM Daniel Cooke notifications@github.com wrote:
The argument is a (CSV) file path where Octopus will write the data profile to, e.g.:
$ octopus -R ref.fa -I reads.bam -o calls.vcf --data-profile reads.profile.csv
You then use one or more profiles to generate new error models https://github.com/luntergroup/octopus/wiki/How-to:-Use-error-models:
$ profiler.py -P reads.profile.csv -O reads.model
Then you can use this error model for variant calling:
$ octopus -R ref.fa -I reads.bam -o calls.vcf --sequence-error-model reads.model
There is no truth set needs to fit the error model.
— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub https://github.com/luntergroup/octopus/issues/85?email_source=notifications&email_token=AMYUHWK5VFO5VKHAG6MCMHDQNWL2FA5CNFSM4I62MDHKYY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGOEAXFACQ#issuecomment-539906058, or mute the thread https://github.com/notifications/unsubscribe-auth/AMYUHWMP56E7BETSVQIXZNDQNWL2FANCNFSM4I62MDHA .
@dancooke , one clarification on this, do the other arguments to octopus matter? E.g. if the data is from haploid clones, would the -C polyclone
and other appropriate arguments matter?
I just want to ask, what arguments to i specify for --data-profile . Does is requires a vcf files or a truth file?