Can anyone help me understand how to interpret copy number variation (CNV) graphs generated from NGS data? I'm particularly interested in the copy number plots shown in Figure 1 of this paper: https://www.nature.com/articles/s41598-020-64335-3/figures/1.
I'm looking for tutorials, books, review articles, or educational resources that explain these graphs in detail, including how to interpret copy number changes, gains, losses, segmental abnormalities, mosaicism, and the underlying principles of NGS-based CNV analysis.
Any recommendations for beginner- to advanced-level resources would be greatly appreciated. Thank you!
1 answer
For something that's actually about reading the plots rather than running a tool, the CNVkit documentation is the best free thing I know of - its interpreting-results section walks through log2 ratio, segmentation and the allele-frequency track with real examples.
The concept that makes those figures click is the second track, B-allele frequency. Depth tells you how many copies, BAF tells you whose. A copy-neutral loss of heterozygosity has entirely normal depth and only shows up as the heterozygous points splitting away from 0.5, so reading the top panel alone you'd miss it completely.
Mosaicism is the same idea with arithmetic attached. The depth ratio scales with the fraction of cells carrying the change, so a full trisomy sits at log2 0.58 while a 30% mosaic one sits nearer 0.2. That's why mosaic calls need much tighter noise than germline ones to be believable.
Log in to answer this question.
Is there something in particular you do not understand? say in reference to the figure you linked?
Generally, these analyses are done by aligning reads to a reference genome, calculating the number of reads covering each base in the genome (or average on windows etc), normalised to the genome wide average (so a coverage of ~1 would be the say as the genome wide average, and for a haploid 2 would mean there is two copies of this region in your sample), and then plotting this normalised coverage along chromosomes (again, generally averaged across windows of say 1kb).
If you have something you particularly don't understand it'd be easier for people to help.
I am new to NGS and would like to understand how CNVs are interpreted from sequencing graphs, particularly in relation to specific mutations. I am looking for a detailed explanation of how to read and interpret these graphs. Could you please recommend or share any tutorials, books, or other learning resources that explain this topic in a clear and beginner-friendly way?
Those figures are incorrect - the y axis point 1.80 should be 0.80. That should help your interpretation of results. Else the 1-2 haploid diploid expected ratios should be pretty easy to understand from basic genetics.