filter-classified-reads 0.2.0

Filter for reads from taxa of interest using Kraken2/Centrifuge classification results

filter_classified_reads

Filter for reads from taxa of interest using Kraken2/Centrifuge classification results.

• Free software: Apache Software License 2.0

• Documentation: https://filter-classified-reads.readthedocs.io.

Features

• Filter for union of reads classified to taxa of interest Kraken2 and Centrifuge (by default filter for Viral reads (taxid=10239))

• Output unclassified reads along with reads from taxa of interest or exlude them with –exclude-unclassified

• seqtk for quickly filtering reads and pbgzip for parallel block Gzip compression of output reads (recommended that these dependencies are installed with Conda)

Usage

Paired-end reads with classification results by both Kraken2 and Centrifuge

filter_classified_reads -i /path/to/reads/R1.fq \
                        -I /path/to/reads/R2.fq \
                        -o  /path/to/reads/R1.filtered.fq.gz \
                        -O  /path/to/reads/R2.filtered.fq.gz \
                        -k  /path/to/kraken2/results.tsv \
                        -K  /path/to/kraken2/kreport.tsv \
                        -c  /path/to/centrifuge/results.tsv \
                        -C  /path/to/centrifuge/kreport.tsv \

Using test data in tests/data/:

$ filter_classified_reads -i tests/data/SRR8207674_1.viral_unclassified.seqtk_seed42_n10000.fastq.gz \
                          -I tests/data/SRR8207674_2.viral_unclassified.seqtk_seed42_n10000.fastq.gz \
                          -o r1.fq.gz \
                          -O r2.fq.gz \
                          -k tests/data/SRR8207674-kraken2_results.tsv \
                          -K tests/data/SRR8207674-kraken2_report.tsv \
                          -c tests/data/SRR8207674-centrifuge_results.tsv \
                          -C tests/data/SRR8207674-centrifuge_kreport.tsv

You should see the following log information:

2019-04-16 13:40:34,114 INFO: Parsing centrifuge results into DataFrame [in target_classified_reads.py:49]
2019-04-16 13:40:34,168 INFO: Parsed n=12281 centrifuge result records into DataFrame from "tests/data/SRR8207674-centrifuge_results.tsv" [in target_classified_reads.py:57]
2019-04-16 13:40:34,172 INFO: Parsed n=298 centrifuge Kraken-style report records into DataFrame from "tests/data/SRR8207674-centrifuge_kreport.tsv" [in target_classified_reads.py:60]
2019-04-16 13:40:34,177 INFO: Found 7129 unclassified reads from Centrifuge results [in target_classified_reads.py:65]
2019-04-16 13:40:34,242 INFO: Found 231 unique viral Taxonomy IDs [in target_classified_reads.py:98]
2019-04-16 13:40:34,245 INFO: Found 2181 target reads from centrifuge results [in target_classified_reads.py:101]
2019-04-16 13:40:34,245 INFO: Parsing kraken2 results into DataFrame [in target_classified_reads.py:49]
2019-04-16 13:40:34,289 INFO: Parsed n=20000 kraken2 result records into DataFrame from "tests/data/SRR8207674-kraken2_results.tsv" [in target_classified_reads.py:57]
2019-04-16 13:40:34,293 INFO: Parsed n=139 kraken2 Kraken-style report records into DataFrame from "tests/data/SRR8207674-kraken2_report.tsv" [in target_classified_reads.py:60]
2019-04-16 13:40:34,295 INFO: Found 1737 unclassified reads from Centrifuge results [in target_classified_reads.py:65]
2019-04-16 13:40:34,325 INFO: Found 26 unique viral Taxonomy IDs [in target_classified_reads.py:98]
2019-04-16 13:40:34,331 INFO: Found 8345 target reads from kraken2 results [in target_classified_reads.py:101]
2019-04-16 13:40:34,332 INFO: Found N=1701 common unclassified reads by all classification methods. [in cli.py:110]
2019-04-16 13:40:34,333 INFO: Total viral reads=8357 [in util.py:37]
2019-04-16 13:40:34,333 INFO: Centrifuge found n=12 target reads not found with Kraken2 [in util.py:38]
2019-04-16 13:40:34,333 INFO: Kraken2 found n=6176 target reads not found with Centrifuge [in util.py:40]
2019-04-16 13:40:34,338 INFO: N=1701 reads unclassified by both Centrifuge and Kraken2. [in util.py:62]
2019-04-16 13:40:34,345 INFO: Writing n=9999 filtered reads from "tests/data/SRR8207674_1.viral_unclassified.seqtk_seed42_n10000.fastq.gz" to "r1.fq.gz" [in cli.py:129]
2019-04-16 13:40:34,957 INFO: Writing n=9999 filtered reads from "tests/data/SRR8207674_2.viral_unclassified.seqtk_seed42_n10000.fastq.gz" to "r2.fq.gz" [in cli.py:134]
2019-04-16 13:40:35,459 INFO: Done! [in cli.py:137]

Credits

This package was created with Cookiecutter and the audreyr/cookiecutter-pypackage project template.

History

0.2.0 (2019-09-23)

• Use seqtk subseq instead of screed for pulling reads of interest from files

• Added external dependencies for seqtk for pulling reads from input FASTQs and pbgzip for parallel block Gzip for output of Gzipped FASTQs

• Removed screed Python dependency

0.1.0 (2019-04-15)

• First release on PyPI.