This is a test version of Biostars. For the public version, visit https://www.biostars.org.
Centrifuge - Error: reads file does not look like a FASTQ file

I am getting the error that my fastq files are not actual fastq files from Centrifuge. Not sure what is wrong with my files. They seem like normal fastq files and have run successfully through other metagenomic pipelines like metaphlan2, but fail with Centrifuge. Here is prep & my command:

nufile=echo $file | cut -d '_' -f 1
if [[ $file == "${nufile}_1.fastq" ]]; then

    echo "Now processing: $nufile"

    centrifuge -x /home/ubuntu/GTDB_r89_54k/gtdb_r89_54k \
        -1 /home/ubuntu/HMP2_MGX/reads/${nufile}_1.fastq \
        -2 /home/ubuntu/HMP2_MGX/reads/${nufile}_2.fastq \
        -q -S ${nufile}_classification \
        --report-file ${nufile}_report.tsv -t -p 32

Here is a sample of my fastq files:

@CAPPMANXX170326:7:1101:10001:90412/1
TTCTAAATGATCAAGACATATTAAATGCTCTATTATACGACCAAAAGGTACTAATAGATATAAAGTGG
+
BBBBBFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFF<FFFF
@CAPPMANXX170326:7:1101:10004:75175/1
GCAGGGAGAAGACCGCTGTATTGTGGCCATCGAAGTGAACGGCGAAGCAAAGAAGTTCTTCACCAACAGCGAGGAAATGAAG
+
/BBBBFBFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFBFFFFFFFFFFFFFFFFFFFFFFFFFF
@CAPPMANXX170326:7:1101:10005:95028/1
CTCTTAACACTTATGAGTGGTTCCCTAAAAATACAGATCCATATCCGGTAGAGGCTGTTGATGCTGAATTTATTACCTCGGATATACTTACTTTGCCTGAT

Here is my error:

Time loading forward index: 00:06:00 Error: reads file does not look like a FASTQ file terminate called after throwing an instance of 'int' (ERR): centrifuge-class died with signal 6 (ABRT) (core dumped) `

Any ideas on what could be causing this problem ?

centrifuge metagenomics shotgun fastq

Shouldn't that be nufile=$(echo $file | cut -d '_' -f 1)?

sure, but thats not the issue. It is in parentheses in the actual code, which I couldn't post here. The issue is that centrifuge is throwing an error about my fastq files

Thanks, will do and will get back to you

Figured it out, Something was corrupted in my reverse reads, going to re-download them. Thanks again!

0 answers

No answers yet.

Log in to answer this question.