QuickTest Primer will aim to display, in realtime, whether your primer will form a hairpin loop or a primer dimer.
The scoring essential counts up hydrogen bonds - 2 for AT, 3 for GC and subtracts 5 where there are mismatches. The “thresholds” section determines what values are considered important. So four consecutive AT base pairs would have a score of 8 which would pass the 3’ End Primer Dimer threshold and also the Hairpin Structure threshold. The scoring uses values from the settings dialog.
These settings are (defaults value is displayed).
For hairpin detection, we have hard-coded settings for the allowable “loop”. It must be at least 3 nucleotides, but no more than 10. We scan every possible combination to find the highest scoring hairpin e.g.
Here there are 4 consecutive matches - TCAC - so the score is 2+3+2+3 = 10. There is a 7 nt loop. You can get more complicated hairpins, where there is a mismatch in the middle of the stem. The default is to subtract 5 for a mismatch, so to extend this particular hairpin, you would need to have the mismatch be followed by at least two paired residues (one of which would need to be a GC pair) to match or exceed the score of 10, e.g.;
Note that the algorithm is fairly simple and does not try to compute bubbles in the stem where you have e.g. 1 residue looped out in one side of the stem and two or more looped out in the other. The thermodynamics of that get pretty complicated and are not well defined. The approach here is simple and fast and generally alerts you to any significant problems you are likely to encounter.
The Primer Duplex score is essentially identical, except that now you are considering two separate molecules rather than the intra-molecular structure of a single DNA strand. The thresholds are simply the minimum score that needs to be achieved before the structure is reported. We have chosen defaults that we think represent values that are at the low end of what should be acceptable i.e. we err on the side of over-reporting potential issues. Practically, you can probably ignore hairpin structures with values less than 12 for most PCR experiments.
On the other hand, this interface can be very useful for finding inverted repeats - select a long (e.g. 120 bp) region of a target molecule, invoke Quicktest Primer, then click the arrows at each end to scroll through the sequence looking for hairpins. Its a great way of looking for potential one-out restriction enzymes too. Now the primer-primer/primer-template alignment doesn’t use these values. It simply looks for perfect matches between the primer and the template, but allows X number of mismatches, again controlled by the settings dialog. TEMPLATE BINDING > MAX MISMATCHES.
Again, mismatched loop-outs are not taken into account i.e. if you have even a single extra residue inserted in the primer relative to the template, the match will not be shown (unless that mismatch is at one end or the other). So, essentially, gaps in the primer will completely prevent alignment with the template. Residues in the “tail” of the primer are not included in the comparison.
If you need to take gaps into account, you may prefer to use the Align to Reference tool, though you’ll need to adjust the parameters for short primer sequences.
How to design primers with tails and mismatches.