Copy number determination using sequencing coverage
Hi! We are trying to figure out plasmid copy number for a bacterial organism (Klebsiella pneumoniae) and this was how we approached it by NGS:
1. Genomic sequencing from same bacterium using MiSeq and PacBio
2. Mapping MiSeq reads onto PacBio scaffolds/assemblies (one 5 mbp chromosome and two 100 kbp plasmids)
3. Calculate relative coverage using number of MiSeq reads mapped to chromosomal vs plasmid scaffolds to determine plasmid copy number
The results from this approach demonstrated copy numbers of the two plasmids as 4 and 3. I am a little concerned with this approach, since we never did anything to compensate for noise or sequencing bias (as suggested in the literature using algorithms like HMM). We also performed the standard qPCR using single copy genes (gyrA and rpoB) as a reference and had a conflicting finding (only one copy for each plasmid).
Do you guys think that my concerns were valid re: using simple calculation of read coverage to determine copy number? Should I even worry about this? Thank you!
1. Genomic sequencing from same bacterium using MiSeq and PacBio
2. Mapping MiSeq reads onto PacBio scaffolds/assemblies (one 5 mbp chromosome and two 100 kbp plasmids)
3. Calculate relative coverage using number of MiSeq reads mapped to chromosomal vs plasmid scaffolds to determine plasmid copy number
The results from this approach demonstrated copy numbers of the two plasmids as 4 and 3. I am a little concerned with this approach, since we never did anything to compensate for noise or sequencing bias (as suggested in the literature using algorithms like HMM). We also performed the standard qPCR using single copy genes (gyrA and rpoB) as a reference and had a conflicting finding (only one copy for each plasmid).
Do you guys think that my concerns were valid re: using simple calculation of read coverage to determine copy number? Should I even worry about this? Thank you!