https://nf-co.re logo
Join Slack
Powered by
I am a little bit confused with `bracken` and `kra...
# taxprofiler
l
I am a little bit confused with 
bracken
and 
kraken2-bracken
in profiling.nf. What’s the difference between them?
j
What do you mean?
And where exacrtly?
l
Copy code
KRAKEN2_KRAKEN2(ch_input_for_kraken2.reads, ch_input_for_kraken2.db, params.kraken2_save_reads, params.kraken2_save_readclassifications)
        ch_multiqc_files = ch_multiqc_files.mix(KRAKEN2_KRAKEN2.out.report)
        ch_versions = ch_versions.mix(KRAKEN2_KRAKEN2.out.versions.first())
        ch_raw_classifications = ch_raw_classifications.mix(KRAKEN2_KRAKEN2.out.classified_reads_assignment)
        ch_raw_profiles = ch_raw_profiles.mix(
            KRAKEN2_KRAKEN2.out.report.map { meta, report ->
                def new_tool = [meta + [tool: meta.tool == 'bracken' ? 'kraken2-bracken' : meta.tool], report]
            }
        )
    }
j
I think we use that in the later standardisation subworkflow
(I'm on my phone atm, so a bit hard to check exactly sorry)
l
No worries!
j
And was something to do with how we disabimguoate
l
Here are command lines for kraken2-bracken
Copy code
#!/usr/bin/env bash -C -e -u -o pipefail
kraken2 \
    --db testdb-bracken \
    --threads 4 \
    --report 2612_ERR5766176_B_db1.bracken.kraken2.report.txt \
    --gzip-compressed \
    --unclassified-out 2612_ERR5766176_B_db1.bracken.unclassified#.fastq \
    --classified-out 2612_ERR5766176_B_db1.bracken.classified#.fastq \
    --output 2612_ERR5766176_B_db1.bracken.kraken2.classifiedreads.txt \
    --paired \
     \
    2612_ERR5766176_B.unmapped_1.fastq.gz 2612_ERR5766176_B.unmapped_2.fastq.gz
Here are commands for bracken:
Copy code
bracken \
     \
    -d 'testdb-bracken' \
    -i '2611_ERR5766174_db1.bracken.kraken2.report.txt' \
    -o '2611_ERR5766174_db1.bracken.tsv' \
    -w '2611_ERR5766174_db1.bracken.kraken2.report_bracken.txt'
bracken
used the database 
db1
and 
kraken2
used 
db2
, we got an empty 
db1
folder in 
kraken2
results folder. Should I turn off the publishdDir for 
db1
in 
kraken2
?
If you want, we could discuss about that next Thursday
j
Yes probably easier to talk then, but i think they makes sense. I think we have a particular flag to save upstream kraken results if you're running bracken? Or no?
l
Is it possible to have a quick huddle this afternoon between 13:00 and 14:15?
j
Good morning! In pinrciple yes, but wouldn't that overlap with the FAIRy hackathon?
I'm happy to meet over lunc htough
l
Sounds perfect. What time do you prefer? I am more flexible
I also registered the hackthon
j
I guess 1230 1300, now the'yve confirmed the correct tiemzone 🙂
l
Sounds good.
Talk to you soon
j
(12:30ish obviously, because lunch 😬 )
l
No problem. Call me whenever you are ready during the lunch.
I am sorry for taking your lunch time
j
No no all fine!
Just saying if I don't turn up exactly on time I'm still cmoing 🙂
(that I haven't forogt)
l
Take your time.
j
There in 5
l
Perfect!
Open in Slack
PreviousNext
Powered by