1
$\begingroup$

I am aligning a fastq file as follows:

cutadapt -j 8 --quality-base 33 -l 20 \
     -o - /home/user/Documents/analyses/CRISPR-screens/screen3/D16.fq.gz | 
       ~/bowtie2-2.4.3-linux-x86_64/bowtie2 --no-hd -p 8 \
         -t -N 0 -x ~/moffat_tko3 - | 
           sed '/XS:/d' | cut -f3 | sort | uniq -c > ~/D16.guidecounts.txt

The expected output is similar to this:

head D16.guidecounts.txt 
329 A1BG_sgA1BG_1
669 A1BG_sgA1BG_2
65 A1BG_sgA1BG_3
397 A1BG_sgA1BG_4
100 A1CF_sgA1CF_1
731 A1CF_sgA1CF_2
396 A1CF_sgA1CF_3
536 A1CF_sgA1CF_4
427 A2ML1_sgA2ML1_1
978 A2ML1_sgA2ML1_2

However, there is no output from this command as bowtie2 gets stuck. cutadapt works fine, but when that step is done, nothing happens for hours and it should only take a few minutes. I have tried to take a subset of reads using head and tail as follows:

zcat D16.fq.gz | head -10000 | gzip -c > D16head.fq.gz
zcat D16.fq.gz | tail -10000 | gzip -c > D16tail.fq.gz

What confuses me is that these files don't freeze bowtie2 and align well (> 90% alignment rates). Other files from the same sequencing run also align fine (NovaSeq SP flow cell 50 bp paired-end).

As far as I can tell the original fastq file looks fine:

zcat D16.fq.gz | head -12
@A00489:1088:HKN2TDRXY:2:2101:1334:1172 1:N:0:GCCAAT
GCCCAGTGACCCTTTCACGGGTTTTAGAGCTAGAAATAGCAAGTTAAAAT
+
FFFFFFFFFFFF:FFFFFFFFFFFFFFFFFFFFFFFFFFFFFFF,,FFFF
@A00489:1088:HKN2TDRXY:2:2101:1606:1172 1:N:0:GCCAAT
AAAGTTCTATCTGTTGGGCGGTTTTAGAGCTAGAAATAGCAAGTTCAAGT
+
FFFFFFFFFFFFFFFFFFFFF:FFFFFFF:FFFFF:FFFFFFFFF,FFFF
@A00489:1088:HKN2TDRXY:2:2101:1624:1172 1:N:0:GCCAAT
GCGACGGTTCATCTCCGCAAGTTTTAGAGCTAGAAATAGCAAGTTAAAAT
+
FFFFFFFFFFFFFFFFFFFFFFFFFFFFFF:FFFFF:FFFFFFFF,FFFF


zcat D16.fq.gz | tail -12
@A00489:1088:HKN2TDRXY:2:2278:30644:37059 1:N:0:GCCAAT
AGGTCATCCACTCTAGACCTGTTTTAGAGCTAGAAATAGCAAGTTAAAAT
+
FFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFF
@A00489:1088:HKN2TDRXY:2:2278:30915:37059 1:N:0:GCCAAT
ATGAGGAGGGTGATGTTCCGGAGTTTTAGAGCTAGAAATAGCAAGTTAAA
+
FF:FFFFFFFFFFFFFFFFFFFFFFF,FFFFFF:FFFFFFFFFFFFFFFF
@A00489:1088:HKN2TDRXY:2:2278:31638:37059 1:N:0:GCCAAT
GAGTGTGGTTGAGATCCCAAGTTTTAGGCTAGAAATAGCAAGTTAAAATA
+
FF:FF:F:FFFFFFFF,FFFFFF:FF::FFFFFFFFFFFFFFFFFFFFFF

Does anyone know what could be wrong?

$\endgroup$
1
  • $\begingroup$ I would suggest breaking up the piped commands and see if you get some errors when the commands are run separately $\endgroup$
    – Bioathlete
    Oct 18 '21 at 17:21

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.