Illumina TruSeq PCR-Free - Low Final Concentration WGS Rat DNA Libraries

[agileta]

We are trying to run whole genome sequencing on ~40 rats for a project. We performed the preps following the TruSeq DNA PCR-free LT protocol for the 550bp fragment size. Upon completion of a few of these libraries, the range of concentrations we received was ~0.3-1.3nM from qPCR, whereas the recommended concentration to move forward with pooling and sequencing is 2nM+. To make sure this wasn't just experimenter error, we sent the samples to the on-campus core to be prepped by the same method and they ended with libraries concentrated mostly in the 0.1-0.5nM range as well. At the same time however, they had successfully prepped a separate set of samples from a different lab that all met the appropriate end concentration. This potentially argues that it is something about our samples that is affecting the library preparation.

In order to isolate this rat genomic DNA, we use a phenol:chloroform:isoamyl alcohol (PCI) extraction, followed by a saturated chloroform clean-up and subsequent ethanol precipitation. At the end, we are left with DNA concentrated at ~100-200ug/uL suspended in milliQ water with 260/280 in the 1.8-1.85 range and 260/230 in the 2-2.3 range. So theoretically, this issue isn't due to any of the obvious contaminants that this quantification would detect in the 230nm spectrum.

Right now we are at a bit of a loss as to why these library preps are coming out so lowly concentrated. I am about to run a gel to check for sample degradation, but we do not believe this will be the case, as many of the tail samples have been collected in the past year and all are stored in ethanol at -80C and were only recently extracted. We contacted Illumina and they led us to this bulletin linked to below discussing potential contaminants that could be present, but most we do not expect to see in our samples, however we can't rule them all out. (http://cancer.osu.edu/research/cance...rep%20Kits.pdf)

Basically, we are looking for any insight as to why we might be experiencing this issue. We are aware that there are ways to use low concentrations and move ahead with the Illumina sequencing, but we would really like to resolve this issue if it is related to the sample preparation, as we have several hundred samples extracted as such and don't want them causing issues on the HiSeq 2500. We are considering trying a DNA purification kit to try and get them even more pure from whatever hidden contamination there may be...specifically Qiagen recommended the PureGene DNA Tissue Kit. We will then rerun the library preps and see if that resolved any issue. Any other advice or suggestions are greatly appreciated, or people who have seen an issue like this before with rodent tail preps or PCI DNA preps. Thanks.

[nucacidhunter]

Presence of ontaminant is less likely cause because after shearing DNA is cleaned with beads and then goes through enzymatic steps. However, trialling a pre-shearing clean up with a column as suggested would be good. Columns and beads are more effective in cleaning particular type of contaminants and using both may provide more effective purification.

You have mentioned DNA concentration of 200 ug/ul which seems to be extremely high. I wonder if you are quantifying 2 ug input DNA with PicoGreen based or spectrophotometric method. Low input definitely will result in low yield. As you have suggested checking DNA sample intactness would be first step. You may also check some samples after shearing to make sure that shear size is within required specification. Low molecular weight DNA will be lost during size selection leaving little material for following steps.

You can concentrate those libraries either by speedy vac before denaturation or reducing elution volume at final bead clean up step. But a library prep which does not yield well may give uneven or biased coverage