Issue with library prep for Ion S5 NGS

[Siya123]

Hello,

I have an issue with library prep for Ion S5 NGS. We check the concentrations of labraries with Qubit fluorometer(1x dsDNA HS kit) and it doesn't detect anything("Too low concentrations") after normalization. However, when mixed the libraries into pool, they seems to be appropriate( ~18ng/ml). The final result was blue chip with loading of 8%. Since we don't have a kit for qPCR in order to check the quality of the libraries, could anybody suggest what we have done wrong? At first we tought that the kit for the measurement doesn't work, but we've switched it with a New one and the result was the same.

[Ben3]

Hello Siya123 What normalization method and library preparation protocol are you using? And before Qubit quantitation, are you performing a library amplification step?

[Siya123]

Hi, Ben3. We use Ion AmpliSeq Library Kit 2.0 for 2 primer pool. The protocol include target amplification, purification with FuPa reagent, ligation of adapters, purification with magnetic beads and library amplification. We measure the concentrations of the already amplified libraries with Qubit, as we follow the protocol, but for some
libraries there is a problem with the quantitation. Ususally If we manage to quantify the library, we use the table in the manual to calculate the dillution factor: for amplicons with approximate size of 250bp, we divide the library concentration by 17ng/mL(~100pM). For example If we have library concentration of 500ng/ml it should be divided by 17, which results in dillution factor 29.4
Therefore, according to the manual, to prepare the pool we need to dilute 1ul of the library with 28.4ul Low TE buffer. And there is another problem with the measurement and results below the limit of Detection. The last time one of the pools has a good concentration(18ng/ml), but poor loading on the chip(8%). We load another chip with the same pool, but instead of 1ul + low TE, we mixed 2 ul of each library + Double volume lowTE(c~20ng/ml). The loading was better, but still not good enough(51%). Since these results I'm not sure If there is a problem with the first Step of target amplification or there is a problem when preparing/calculating the pool. Any idea?

[Ben3]

Hello Siya123 ,

Thanks for the additional information. The recommendations for the dilution factor are starting recommendations, but may have to be optimized based on your actual sequencing results. Occasionally, the panels or samples may not perform optimally. This is likely due to sample quality, quantity issues, or other factors that may affect amplification. If you have a control DNA sample, you can try running that in parallel to your own sample, to test whether it is an issue with your starting material.

It's possible that you have to increase the amount of library you load for these types of samples/experiments. I would recommend trying 2uL of library + low TE, or 3uL + double volume low TE, and see if that increases your loading percent.

Let me know what you end up trying and if you end up fixing the problem. Thanks!

[Siya123]

Hi @Ben3,

Thank you once again for your suggestions. As I previously wrote, we've already tried to increase the volume of the libraries: instead of 1ul + low TE, we mixed 2 ul library + double volume low TE. As a result we have 8% loading(1ul library) and 51% loading(2ul).May be next time we would try your suggestions for even bigger volume. Also it's a good idea to use a control.
Unfortunately there is no control DNA. We have a lot of good runs, but the starting material has been isolated by different methods: either from Trizol reagent or by DNA columns. We have issues with both of them - the Trizol DNA is really hard to be dissolved(it takes a lot of time and it's not optimal) and recently we had an issue with the columns - some of them resulted in none DNA isolated. So as you can see there are a lot of problems.
According to the dillution factor, at the beggining we started by dividing our concentrations by 15(according to the manual), and passed through 16 and 17, since the last one was optimal for our run.
Thank you once again for you suggestions.