Skip to content

Latest commit

 

History

History
79 lines (55 loc) · 2.42 KB

101Kraken2.md

File metadata and controls

79 lines (55 loc) · 2.42 KB
Raw

Metagenome data analysis workflow by Kraken2

Download annotated FASTA for Kraken2

Annotated FASTA file for Kraken2 (16S_kraken2.fasta) is proved by Life Science Database Archive at JST National Bioscience Database Center (NBDC). Please download from the download site of 16S RNA sequence set for metagenomic analysis.

Building library

1. Set working directory

    $ DBNAME="/home/{yourname}/kraken2/db/"

2. Downloading required file(taxonomy and so on: names.dmp and nodes.dmp)

    $ kraken2-build --download-taxonomy --db $DBNAME

Docker version

kk2 is the name of kraken2 directory in Docker environment.

    $ docker run -it -v `pwd`:/kk2 staphb/kraken2 kraken2-build --download-taxonomy --db /kk2/db

3. Installing reference library

    $ kraken2-build --add-to-library 16S_kraken2.fasta --threads {n} --db $DBNAME

Docker version

    $ docker run -it -v `pwd`:/kk2 staphb/kraken2 kraken2-build --add-to-library /kk2/16S_kraken2_db_2021_5b.fasta --threads 14 --db /kk2/db

You may see error message for dustmasker. We now ignore this error.

Masking low-complexity regions of new file...Unable to find dustmasker in path, can't mask low-complexity sequences

4. Building database

Specify {n} for the number of threads to use.

    $ kraken2-build --build --threads {n}  --db $DBNAME

Docker version

    $ docker run -it -v `pwd`:/kk2 staphb/kraken2 kraken2-build --build --threads {n} --db /kk2/db

Classification & Visualization

1. run kraken2

FASTQ file (sample1.fq.gz) in fq directory. The classification report will be written to report.txt.

    $ kraken2 --db $DBNAME --gzip-compressed fq/sample1.fq.gz --threads {n} --report report.txt --output -

Docker version

    $ docker run -it --env KRAKEN2_DB_PATH=`pwd`/db -v `pwd`:/kk2 staphb/kraken2 kraken2 --db /kk2/db --gzip-compressed /kk2/fq/sample1.fq.gz --report /kk2/F2s.txt --output -

2. visualization

This part is not included in shell script because multiple results can be included and compared in this visualization. So, run the commands below separately.

$ pip install jinja2 pandas
$ python kraken2_report_formatter.py -n sample_name_1 sample_name_2 -f sample_report_1 sample_report_2