I have one file of Illumina sequencing reads in which the paired-end reads appear to be merged. Each sequence in the file is 72 bp in length even though the run yielded 36 bp reads with 163 bp max insert size. They do not appear to be tab delimited. This is different from my past experience, in which paired ends are always in two different files.

Each read looks like this:

@DRR000001.2 3060N:7:1:1114:186 length=72
GATATTGGCCTGCAGAAGTTCTTCCTGAAAGATGATGGAAACCTCAGGGGCCGAATATCGGAGAGCAAAAAG

My question is: how do I run this in Bowtie? The following gives me an error:

bowtie indexfile -12 trimmed.fastq pair.sam --al pair.align --un pair.unmapped -l 24 -X 200 --fr -q -t --sam -n 3

because -12 should be two files or a tab delimited file

How do I get this to work?

A little extra info: I got these files from NCBI's SRA database

Now I have a different problem. Here is my code:

./fastq-dump --split-files ~/DRR000001.sra
./fastq_quality_filter -Q33 -q 25 -p 70 -i ~/DRR000001_1.fastq -o ~/natto1_trim.fastq
./fastq_quality_filter -Q33 -q 25 -p 70 -i ~/DRR000001_2.fastq -o ~/natto2_trim.fastq
bowtie-build -f bsubtilisnatto.fasta bsubtilisnatto
bowtie bsubtilisnatto -1 natto1_trim.fastq -2 natto2_trim.fastq pair.sam --al pair.align --un pair.unmapped -l 20 -X 200 --fr -q -t --sam -n 3

Almost nothing will align. I lowered my thresholds down to seedlength=20 and mismatches = 3, and also tried --ff. What's not working?