The score in the output BEDfile from the primersearch process currently represents a 'raw' score of the nuc44 scoring matrix combined with custom optimized gopen/gextend values.
This results in the score not being all that useful for the primersearch process as it becomes unreasonably hard to quickly understand if a found primer aligned well or not based on a score of 95, for example. Especially because we're aligning, and scoring, very short sequences.
It's probably better to represent this in another metric, like a percentage or edit distance.
i.e. when using a percentage, if the primer has a length of 30bp and the 'score' is 90. Then it's easy to determine that at least 27 of the 30 primer nucleotides were a match.
The score in the output BEDfile from the
primersearchprocess currently represents a 'raw' score of the nuc44 scoring matrix combined with custom optimized gopen/gextend values. This results in the score not being all that useful for the primersearch process as it becomes unreasonably hard to quickly understand if a found primer aligned well or not based on a score of 95, for example. Especially because we're aligning, and scoring, very short sequences.It's probably better to represent this in another metric, like a percentage or edit distance. i.e. when using a percentage, if the primer has a length of 30bp and the 'score' is 90. Then it's easy to determine that at least 27 of the 30 primer nucleotides were a match.