Low Pass Filter %

[bwking82]

Hello All,

I work in a core sequencing facility (we do not perform Library preps) and we are having a problem with a customer's samples. It's actually 2 problems as we are having an issue with cluster density as it relates to loading concentration (normally loading 5 pM works very well but in this case it only generates ~ 30,000 clusters per tile). We quantitate using Qubit fluorencense and the Experion to get our size range correct and both returned expected results. The other issue, and these may be related, is that the pass filter % is very low (ranging to 70% to as low as 25%) and we observed the run as it took place and did not notice an abundance of clusters that would indicate overloading.

Does anyone have any experience with either issue and any methodology that could be used to remedy these problems. The client is getting anxious about his samples but I am wary of wasting more flow cell lanes until we can adress this issue. The sample is a yeast with a fairly balanced GC content.

Brian

[elaney_k]

Hi Brian,

Have you tried to quantify the library using a QPCR assay similar to that published by the sanger centre (Nat Methods. 2008 Dec;5(12):1005-10)? I've found that the results using the QPCR validation correlate much higher with cluster number than the qubit measurements and you can spot libraries that are going to give you lower cluster numbers before loading onto the flowcell so you can load more of these samples to compensate.
What does the alignment rate with your reference genome look like? If it's lower than expected there may be some adapter-dimers in the library, which is possible if the customer didn't titrate the adapter ratio down to match the input quantity of DNA.

Elaine