high sequence duplication ddRAD

Question

[gubrins]

Hello,

I'm relatively new to NGS analyses. I'm working with single-read ddRAD data of a non-model species and we just obtained the fastq files. The company removed the adaptors and I've just did the demultiplexing and some trimming, to remove the overhangs. When I run FASTQC and MultiQC, I obtain a high degree of duplication (around 80%). I've seen that this could be normal in RNA-seq data, but what about ddRAD? As I just started handling the data I don't think I did something wrong, but I find this high number of duplications really strange. What do you think?

Thanks in advance :) https://ibb.co/Y2LY4VH

[GenoMax]

Have you checked to see how ddRAD works? If not start here and take a look at some of the papers included in that link.

[gubrins]

Thanks for answering. I'm aware of how ddRAD works, but what I don't understand is the pattern I observe in my data. I uploaded a picture, let's see if you can check it.

Answer 1

[GouthamAtla]

If you understand ddRAD, you will know why there is high duplication rate. The restriction enzymes (double digestion) cut at specific positions in the genome, and your library is enriched for those specific fragments (and size selection + PCR amplification). so you tend to sequence same genomic DNA more often than compared to whole genome sequencing methods (in WGS, the fragmentation of DNA is random, so you sequence random fragments more often).

As suggested, read the relevant papers and check how much duplication is reported and how they deal with it.

[gubrins]

thanks to both of you, obviously I don't understand it as I thought. I'll check the papers!