There can be several reasons why your DNA library concentration is not sufficient for sequencing, despite having good genomic DNA concentration. Here are some possible causes and suggestions to address them:

Quality of genomic DNA: It's important to have high-quality genomic DNA for library preparation. Poor quality DNA can lead to low yield and fragmented DNA that may not be suitable for library preparation. You can assess the quality of your DNA by agarose gel electrophoresis or using a spectrophotometer.

Variations in the library preparation protocol: Variations in the library preparation protocol, such as different buffer conditions, can lead to variations in library yield and quality. Ensure that all steps in the library preparation protocol are followed consistently, and consider using validated protocols.

Variations in DNA input: Variations in the amount of input DNA can also lead to variations in library yield. It's recommended to use consistent amounts of DNA in each sample and optimize the input amount if necessary.

Sample degradation: Sample degradation can lead to the loss of DNA quality and quantity. Ensure that samples are stored and shipped under optimal conditions to minimize degradation.

Contamination: Contaminants such as salts, detergents, and inhibitors can interfere with the library preparation and sequencing processes. Make sure that all reagents used in the library preparation are free of contaminants, and consider using a DNase treatment to remove contaminants.

It may also be helpful to consult with the VANTAGE team at Vanderbilt or with a sequencing specialist to troubleshoot the issue. They may have additional suggestions or modifications to the library preparation protocol that can improve your results.

Hello..my apologies if this doesn't apply as I do not know all the steps in the protocol you followed...but two major issues I've come across when using these kinds of kits for gDNA purification is how well (or how effective) the treatment with RNAse is and the actual elution of higher molecular weight DNA from the column. Some sort of bioanalyzer or equivalent might show that there is a fair amount of other stuff in your elution and perhaps a lot smaller than expected.

I've also found that once you load your column, wait a few minutes for the column solution to "activate" the silica so that the absorption and subsequent elution becomes more efficient. The second biggest problem is the guanidine carryover...and the column washes do not efficiently remove the salt which can cause artificial increases in apparent DNA concentration. Similar to the loading, once the ethanol washed are done, I usually wait a few minutes (3 to 5) for better wash and clean up, and I do 3 ethanol washes to maximize the cleaning of guanidine from the system.

Some possible steps to try:

1. Increase the time you perform your proteinaseK step...this may help in the purity issues (as mentioned in the other answer)

2. The RNase treatment should be done after the proteinaseK, which necessitates a heat inactivation step. Ideally, this should be 95 degrees for 10 minutes, but you will shear your DNA....an acceptable compromise is to heat inactivate at 70 degrees for at least 15 minutes...and then add RNAse and incubate at 37 for 30 mins.

3. When you add the sample to the column, wait a few minutes before centrifuging, and centrifuge at 3000-5000g...not the maximum speed. This will aid in the binding to the column. The rest of the washes and elutions can be done at normal speed (>= 10000g).

4. When doing the last ethanol washes (AW2), wait 3 minutes before spinning, and you can do an extra AW2 wash to make sure there are no salts left.

5. For elution, pre-warm your TE (or AE) to 65 degrees...this helps in increasing the elution from the column.

For full disclosure, these steps may not fix your issues, but it may be worth testing to see if they increase your overall success rate, even though I concede that that the time it takes to process your samples will be longer.

Hope this helps!

Nicolas Kuperwasser Thanks a lot. Some of these points are reflected in Qiagen kit public protocol, but you gave me a few things to change. Qiagen's tech support also suggested increasing the time/concentration of proteinase K and adding additional ethanol washes. I will also try your pre-warming elution buffer idea.

Do you have further thoughts on RNase? I'm getting conflicting ideas as I research. Some say it is important but suggest heat treating the RNase to be absolutely certain that it has no DNase activity. Others say that this will likely destroy the RNase activity as well.

Do you have any suggestions for the best way to analyze the DNA to differentiate between RNA contamination, guanidinium salts, and other protein contamination?

Hi...almost all the RNAses that you can find commercially are DNAse-free, and boiling it for a few minutes will indeed inactivate any potential DNAses that are present...it may slightly decrease the activity of RNAse, but not enough to make it not work....it's a tough little bugger, and you can certainly heat-treat the RNAse before using. Also, looking at the absorbance values and the ratios can give some insight into the quality of the prep...and there is always electrophoresis to see if there are smears and the like).

Given the volumes and quantities of material that you have, I don't have an immediate diagnostic assay to differentiate DNA from RNA, other than enzymatic which brings it back to the RNAse step. However, the key point of the RNAse, especially for solid phase purification, is to reduce the size of the fragments so that they do not co-bind with the larger DNA fragments, which is why it would be preferable to do that before loading the columns, but after the proteinasK step (which would inactivate the RNAse in the first step). I use commercial RNAse all the time and have never had an issue with DNAse contaminations.

As far as potential residual protein (if it is there), you could be "overloading" the column by having too much stuff in your original sample when you add the first buffer, and they do not get removed during the subsequent washes. I don't remember the Qiagen protocol off-hand, but basically it's 1 volume sample + 1 volume lysis/binding buffer. In this case, I would add 1 volume of water (or suitable buffer) to your 1 volume of sample and then add 2 volumes of lysis/binding buffer (and proteinaseK) and continue with the protocol....of course you would have to load the column twice (load/spin..and load again and spin), but you would then be able to have enough chaotropes/detergents/etc that prevent protein binding to the column (and perhaps even enhance the recovery of gDNA) and in theory would have a cleaner prep. Because the buffers are concentration dependent, the total amounts of salt will remain the same, regardless of how many times the column is loaded, and they will be removed by the subsequent washes (and as discussed before).

Hope this helps and happy to answer more questions!!

NK

Here is an update on what I have found so far.

Qiagen tech support suggested that I was overloading the Blood MiniKit. I used their protocol for peripheral blood but bone marrow blood has a much greater cell concentration. The overload means that proteinase will be too diluted and not have enough time to fully break down the hemoglobin and the heme components. The iron ions in the elute will interfere with PCR activity. Diluting the sample, adding more proteinase, and increasing the proteinase incubation time should remedy these issues.

Initially, we thought maybe EDTA was interfering with the process, so I used the Qiagen DNA MicroKit to collect the DNA and elute it in another buffer. I did the ethanol, AW1, AW2, and elution steps for 48 samples. Only two of these samples failed to make good libraries despite having very low DNA concentrations (median of 7 ng/uL and quite a >1 ng/uL).

RNA contamination came up the most frequently. I can not rule it out as a possibility, but with multiple freeze/thaw cycles and handling, I doubt that RNA is stable enough to be the primary issue.

Other detergents or salts from buffers could be an issue if they are not getting washed out properly.

Here are the protocol adjustments I plan to make:

Looks good....and good luck!