How to fix fastq files from a bacteria sample?

[ymc]

I have a pair of PE read files generated from a Truseq library. It has 33M pairs of 150bp reads. The bacteria is S auerus which has a genome size of 2.8Gbp. My goal is to call protein coding variants.

I am getting Per Base Sequence Quality, Sequence Duplication Level and Kmer Content errors from fastqc.

Does this look like it is only because of over-sequencing such that I don't need to do anything and let picard remove duplicates for me after mapping? If not, can you tell me what else I need to do to fix up the fastq.

Thanks a lot in advance!

[GenoMax]

I would say go ahead with the analysis and then see if you have a real problem once you have alignment results.

[Brian Bushnell]

It looks like you have over 3000x coverage, so it's not surprising that some reads appear to be duplicates. You should not deduplicate them unless the library was amplified. But even then, at such a high coverage level, I don't think deduplication is wise because there will be so many true duplicates that occur just by random chance.