Entering edit mode
10 months ago
aussie_roo
•
0
Hi, I'm new to sequencing. I don't understand why most labs set this metric and how they derive these specific values, rather than 95%, or 30X for example. How do they tie in with statistics as well? Does it have to do with an error rate of 1%? Or does it have to do with minimizing FP/FN? How does 20X give statistical confidence and what are the mathematical calculations that prove this?
Also how does this relate to the context of rare disease screening and SNP/CNV/INDEL detection? Thanks
The context of your question is unclear. Who uses 95% (of what?) at 20X and for what? what kind of analysis are you trying to do? Please edit your question and clarify.
I think they are asking 'why is 99% genome coverage at 20X' the standard. What's missing is the target organism and the sequencing technology. For viruses I've seen 90% at greater than 100X was used as a cutoff. It depends on the organism of interest and the question you are asking.
Yes, forgot to mention, it's 99% coverage at 20X depth. The target is human DNA (germline mutations) and on the Illumina Nextseq2000. Thanks for pointing out.
I suggest you read the literature on best practices eg: https://genomemedicine.biomedcentral.com/articles/10.1186/s13073-020-00791-w
Many times, such numbers don't have strict definitions are empirical observations between what is feasible and practical while keeping the numbers in the realm of something also easy to remember like 99% and 20x