We are attempting to automate double-SPRI size selection, in which two successive SPRI-bead-based size cuts are used to select DNA approximately between 200 and 450 bp in length. We are having a very hard time getting the reaction to work and would really appreciate some advice.
Design:
At least in theory. In reality, the amount of size selection we get is minimal, on the order of maybe 10-20% reduction in the high and low bands.
We have experimented heavily and modified the following factors:
None of these has improved size selection. We've exhausted our own ideas for what could be causing the problem. Has anyone encountered similar difficulties before, or is there an obvious factor here that we're missing?
Design:
Our design adds Agencourt Ampure SPRI beads to the DNA at a volume ratio of 0.55, incubates and separates out the beads, then moves the supernatant from that binding reaction into a second reaction where the ratio is increased to 0.7. After the second incubation, we separate, rinse, and elute from the beads. The first reaction cuts out the material above 450 bp, and the second cuts out the material below 200 bp.
At least in theory. In reality, the amount of size selection we get is minimal, on the order of maybe 10-20% reduction in the high and low bands.
We have experimented heavily and modified the following factors:
- Robotic vs manual pipetting
- Different technicians
- Different samples, including unselected DNA libraries and DNA ladders
- Different lots of each reagent, including Ampure beads, ethanol and elution buffer
- Different SPRI : DNA ratios for each cut point
- Incubation periods at 5, 10, 15 minutes
- Smaller and larger volumes
- Different separation magnets
None of these has improved size selection. We've exhausted our own ideas for what could be causing the problem. Has anyone encountered similar difficulties before, or is there an obvious factor here that we're missing?
Comment