trimming primers using iVar

Question

[binfo.sl]

Here is the output after running ivar trim on a bam file.

**Trimmed primers from 25.36% (209222) of reads.

4.84% (39904) of reads were quality trimmed below the minimum length of 60 bp and were not written to file.

71.71% (591672) of reads that started outside of primer regions were not written to file

82.3% (679059) of reads had their insert size smaller than their read length**

I used the command:

ivar trim -m 60 -b {primer-bed-file} -p {output.bam} -i {input-bam}

When I use -e options with above command, the reads that starts outside of primer regions are written to the bam file and hence there is not much change in the depth of bam before and after trimming. But when I do not use -e options, the depth reduced to 1000 from 5000 which makes sense as so many reads are being discarded.

I tried merging the paired end reads and trim the primers. There is not much difference using merged reads in terms of the reads outside of primer regions.

My question is,

1. what difference does this make if we just exclude these reads vs including the reads. Do the reads starting outside of primer regions not map to reference? Is there a way to isolate those reads?
2. Since 71 percent of reads are outside of primer regions based on this output, does this mean we are having quality issue with the sequencing?

Thanks

[Brian Bushnell]

It might be helpful if you could post what you are trying to accomplish and why. And what kind of data you have, the screen output from your command line, the file sizes, etc. The more information available, the easier it is to troubleshoot.

[binfo.sl]

My apologies.

I am analyzing SARS-CoV-2 data sequenced with Artic 5.3.2 primers on an Illumina MiSeq using 'Illumina COVIDSeq Assay'. I suspect it is due to the fragmentation during library preparation that generated reads without primers.