/summarize_kraken_data

scripts to summarize output from Kraken, the taxonomic sequence classifier

Primary LanguageShell

Summarize Kraken Data

emphasis on contamination assessment of bacterial genome sequencing efforts

alt tag

Scripts:

  • summarize_kraken-report.sh -- generates a TSV summary of unclassified, the 3 most abundant genera, and 3 most abundant species from a kraken-report file. (requires GNU sed)

    Step 1 ) Generate kraken report: kraken --threads $CPUs --preload --paired --fastq-input sample_R[12].fq | kraken-report > sample.kraken-report.tsv

    Step 2 ) Summarize: summarize_kraken-report.sh sample.kraken-report.tsv > sample.kraken-summary.tsv

  • summarize-kraken.sh -- makes tab-delimited summary of kraken reports (GNU sed required)

    Step 1) Batch create kraken reports: mkdir ../kraken && for F in *.R1.paired.fq; do B=$(basename $F .R1.paired.fq); kraken --threads 31 --preload --paired --fastq-input "$B".R1.paired.fq "$B".R2.paired.fq | kraken-report > ../kraken/"$B"_kraken.tab; done

    Step 2) Summarize all reports into a single file: for F in ../*_kraken.tab; do summarize_kraken.sh $F >> ../Summary_kraken.tab; done

    Step 3) Append header (optional): awk 'BEGIN{print "ID\tUnclassified Reads [%]\tUnclassified Reads [#]\tunclassified\tTop Classified Reads [%]\tTop Classified Reads [#]\tTop Classified Genus\t2nd Top Classified Reads [%]\t2nd Top Classified Reads [#]\t2nd Top Classified Genus\t3rd Top Classified Reads [%]\t3rd Top Classified Reads [#]\t3rd Top Classified Genus\tTop Classified Reads [%]\tTop Classified Reads [#]\tTop Classified Species\t2nd Top Classified Reads [%]\t2nd Top Classified Reads [#]\t2nd Top Classified Species\t3rd Top Classified Reads [%]\t3rd Top Classified Reads [#]\t3rd Top Classified Species"}1' ../Summary_kraken.tab | cut -f 1-3,5- > ../Summary_kraken-inclheader.tab

Example Installation of Kraken:

# Compile
curl https://ccb.jhu.edu/software/kraken/dl/kraken-0.10.5-beta.tgz -o ~/Downloads/kraken-0.10.5-beta.tgz
mkdir ~/kraken
tar zxvf ~/Downloads/kraken-0.10.5-beta.tgz -C ~/kraken --strip-components=1
bash ~/kraken/install_kraken.sh ~/Kraken
rm -r ~/kraken

# Download MiniKraken DB
curl http://ccb.jhu.edu/software/kraken/dl/minikraken.tgz -o ~/Kraken/minikraken.tgz
mkdir ~/Kraken/MiniKrakenDB
tar zxvf ~/Kraken/minikraken.tgz -C ~/Kraken/MiniKrakenDB --strip-components=1

# Update paths
echo 'export PATH="$PATH:$HOME/Kraken"' >> $HOME/.bash_profile
echo 'export KRAKEN_DB_PATH="$HOME/Kraken/MiniKrakenDB"' >> $HOME/.bash_profile
echo 'export KRAKEN_DEFAULT_DB="$HOME/Kraken/MiniKrakenDB"' >> $HOME/.bash_profile