Saliva DNA used for Nimblegen EZ target capture, resulting in inaccurate CNV estimate

[Cindy007]

Hi all:
We deal with some DNA that are extracted from human saliva and blood specimen for target region capture(Nimblegen EZ choice custom panel) sequencing (illumina CN500 platform). The A260/280 ratios are good as are the nucleic acid concentrations. In our experiment we hybrized and sequenced DNA extracted from blood and saliva of the same person at the same run.
However, the quantitative metrics of saliva DNA such as raw data, average depth,GC%, counts of copy number variations (CNV) varied from the matched Blood DNA . Does anyone has any experience?


Raw data GC% Average depth(whole region)
gDNA extracted from blood 0.86G 42.07 140.44
gDNA extracted from saliva 0.4G 52.84 379.76

We all know the majority of DNA in saliva comes from white blood cells ,however, human saliva also contains bacteria，When extracting DNA from saliva, some of that bacterial DNA is recovered along with the human DNA. Dose the bacterial DNA in saliva have any impact in the result? Why?
Waiting for your feed-back. Thanks a lot!

[carolined]

Hi Cindy,
I work at Roche NimbleGen in the complaint handing department. Based on the information that you have provided, my suspicion is that the DNA from saliva also contains other DNA in addition to the gDNA that you were after, depending on how you isolated the DNA that is. If you also collected bacterial DNA, then this would have reduced the amount of target gDNA that went into the sequence capture and could explain the lower coverage of the saliva samples compared with blood samples. You also might want to try to quanitfy the DNA concentrations using a DNA binding dye, which is more accurate than A260.
If you would like to troubleshoot this further, then please contact your local Roche technical support for assistance. You can find them at http://www.technical-support.roche.com/default.aspx

Caroline