Bowtie2 Alignment for breadth of coverage using -U and --no-unal taking a long time

Question

[taffners]

I'm have been trying to follow the post below to calculate breadth of coverage. However, the alignment step appears to be not working.

http://www.metagenomics.wiki/tools/samtools/breadth-of-coverage

I'm trying to run the alignment with the following command but it has been running now for 1.5 hours on 20 processors.

bowtie2 -p 20 --no-unal -x /test_micro_pipeline/reference_sequences/Escherichia_coli/Escherichia_coli_O104_H4/Escherichia_coli_O104_H4 -U SRR292770_1.fastq,SRR292770_2.fastq -S test_mapping_2/Escherichia_coli_mapping/Escherichia_coli_O104_H4/Escherichia_coli_O104_H4_mapped.sam

I ran bowtie2 without --no-unal and using paired end mapping and this only took about 30 mins but my coverage was really low when I followed the rest of the steps in the metagenomics article. Maybe I'm screwing up something later in the pipeline but my question is why does the command above take so much longer to run than without --no-unal and -U? Also why are the reads being considered as single read?

[ATpoint]

Option -U is used only for single end reads. To run it paired-end you need:

bowtie2 idx -1 read_1.fastq.gz -2 read_2.fastq.gz