Consistent clustering between different pools?

[Beckford_Jay]

Hi All,

I have a question related to cluster density as it relates to getting consistent results between different runs. I am doing RNA-seq using Miseq v3 and when I first started, I got cluster densities around 700K and 950K from loading my DNA samples at 13.5pm for my first two runs. I tried bumping up to 15pm loading concentration to increase cluster density and my cluster density dropped (450K). This is confusing since loading more DNA have should lead to higher cluster density.

From that, I can reasonably infer that loading my pool at 13.5pM for my last run would have given me significantly lower cluster density than what I got from the first two runs. I ran a different set of libraries for each run but the prep kit, quantification and pooling strategy were the same. Does anyone have experience with different libraries/pools generated and quantified using the same exact protocol clustering at different densities even when loaded at the same concentration?

[GSviral]

Hi,

It's tough to get an idea of what is happening just by cluster density and loading concentration alone.

- For the 450 K/mm2 run, what were your pass filter and Q-scores looking like compared to the two successful runs?
- What type of samples are you using and what sequencing are you performing (amplicon, WGS etc.)?
- Are you spiking in PhiX to each of your runs? If so, at what %?
- Have you taken a look at the raw cluster image files? Sometimes, if the MiSeq is overclustered the machine has a hard time defining clusters and it can return a density number which isn't indicative of the actual clustering.
- What is your quantification method? Qubit + fragment analysis?

It would be cool if you could even supply a screenshot of the summary and metrics tab from SAV to further diagnose the problem.