gravatar for v.berriosfarias

2 hours ago by

Hello, I'm trying to align metagenomic read sequences against a reference set of contigs, the samples are each one splitted in two files (these are paired-ended reads) and I'm trying to make a loop using bowtie2 and at the same time making directly a BAM file without making a SAM file on my directory using pipe |

this is my command:

for i in ls /mnt/path/DEC2017/*1.fastq.gz 
do dir=/mnt/path/DEC2017/ 
base=$(basename $i _1.fastq.gz) 
bowtie2 -p 4 -x testbuild -1 $dir/${base}_1.fastq.gz -2 $dir/${base}_2.fastq.gz | samtools view -b -o $dir/${base}.bam -
done

and here is the output (which is repeated a couple of amounts of times)

    stat: Bad file descriptor
Warning: Could not open read file "/mnt/path/DEC2017/ls_1.fastq.gz" for reading; skipping...
Error: No input read files were valid
(ERR): bowtie2-align exited with value 1
[main_samview] fail to read the header from "-".

and the console stills processing this output (as I still see the keyboard pointer symbol, seems frozen.)

So I understand that bowtie2 is not reading any of my files, instead of that is trying to read a file called ".fastq.gz" which clearly doesn't exist.

Here is the format of my files:
DEC2017_L3S3D05_1.fastq.gz
DEC2017_L3S3D05_2.fastq.gz
DEC2017_L3S1D15_1.fastq.gz
DEC2017_L3S3D05_2.fastq.gz
DEC2017_L1S2D05_1.fastq.gz
DEC2017_L1S2D05_2.fastq.gz
and so on...

The problem seems to be in the basename command part, but I can't figure out what exactly is.

I know that this issue is pretty similar to this one to run bowtie on multiple fastq file , In fact, I made this loop taking as reference a command described there but given my lack of knowledge of how exactly basename really works, I cannot identify the exact problem.



Source link