Primer design and BLAST E value stringency

Question

When searching for misprimings I have been told that an e value that is higher than 0.01 is ok and will produce no significant amounts of mispriming. Yet I searched some and it would seem that the e value threshold limit is dependent upon the "amount of information" in the database. Indeed I found that some papers cite as values lower than 0.07 to be significant already.

What are the criteria you use at times to judge this issue, and how do I know when to lower or increase my stringency when running a routine BLAST for the purposes of designing primers?

Answer 1

I never used BLAST for primer design, but often for finding out where a sequence came from.

From the FAQ of the NCBI: http://www.ncbi.nlm.nih.gov/blast/Blast.cgi?CMD=Web&PAGE_TYPE=BlastDocs&DOC_TYPE=FAQ

"The lower the E-value, or the closer it is to zero, the more "significant" the match is. However, keep in mind that virtually identical short alignments have relatively high E values. This is because the calculation of the E value takes into account the length of the query sequence. These high E values make sense because shorter sequences have a higher probability of occurring in the database purely by chance. For more details please see the calculations in the BLAST Course.".

What they say in the papers is true, the size of your database matters in deciding how significant the match is. I can imagine that for primer design you want to be more careful with the E-values, as they often fall under: "because shorter sequences have a higher probability of occurring in the database purely by chance".

I hope this helps to make your decision for deciding your threshold, otherwise please give more information.

Answer 2

This NCBI page offers a primer design tool which appears to incorporate a specificity check run via BLAST. However it seems to base stringency upon degree of mismatch rather than an e value.