PCR-Blast?! - (Nov/15/2004 )
Hello! Can somebody tell me how I can make a BLAST concerning the primers only. I want to use genomic DNA as template and therefore need infromation about primer-specifity. Using BLAT was not quiet sucessful.(http://www.genome.ucsc.edu/cgi-bin/hgPcr)
The PCR does not look good at all, even after optimization-steps. What can I do to find more specific primers? 
Hi
You could try using the BLAST option on the Genbank website (http://www.ncbi.nlm.nih.gov/BLAST/), the Nucleotide-nucleotide BLAST (blastn) is the one you want. Just plug your primers into the box at the top of the page and select your organism from the list below. This will give you a list of the sites where your primers bind and how specific they are.
If you need to design more primers, there are plenty of free sites on the web. I have used primer3 (http://frodo.wi.mit.edu/cgi-bin/primer3/primer3_www.cgi) pretty successfully in the past
Good luck
don't know exactly what you mean. Actaully I think the virtual PCR tool is very useful. you just input your primers and the program will return matches, among them, some are of the expected size, some are not. If there are more than one matches which are the same size as your targeted amplicon, your primers are not specific enough.
If you use NCBI blastn, choose the "Search for short, nearly exact matches" and put some "n"s between your two primers.
I try to explain with the following example: I work with human genomic DNA.
FOR 5' aatgattgtagcaacatcc
REV 5' tttaactttcccagctgcc
I want to ampify AF035968 2536-3021(Chr 5). But I have problems with contamination and wrong amplification and therefor tried to analyse the sequence of the wrong product.
The result was that Chr 6 AL590482 74055-74416 is amplified because parts of the primer bind non-specific (FOR: 12 of 20 bp/ REV: 15 of 20bp). Furthermore there are up to at least 3 more positions for the REV-Primer to bind non-specific (74421-74533). The Blast did not show any of this possibilities. Do I have to convert the REV Primer bfore Blast or can I use the reverse 5'-3' sequence?
Perhaps I use the program in an impropper way!?
Thanks, nabla
The in silico PCR returns a single match on ITGA2 gene which is the same gene as AF035968.
in silico PCR result:
>chr5:52383069+52383341 AATGATTGTAGCAACATCC TTTAACTTTCCCAGCTGCC
AATGATTGTAGCAACATCCcagacatcccaatatggtggggacctcacaa
acacattcggagcaattcaatatgcaaggtaagttttggtgctaataggc
caatgttttcataatgtaaaacattatatttatgtaataaatatgaaaaa
gtaaggaaaagacaaagaaaaataatatacctggtacctaatttaaatca
gaactaataaagaaaaaaacatcagagcattctatgtcttgaatactttg
agaaGGCAGCTGGGAAAGTTAAA
This matched region aligns to AF035968 between 2749-3021
||||||||||||||||||||||||||||||||||||||||||||||||||||||||| ||
Sbjct: 2749 aatgattgtagcaacatcccagacatcccaatatggtggggacctcacaaacacatttgg 2808
integrin alpha 2 36 M I V A T S Q T S Q Y G G D L T N T F G
Query: 61 agcaattcaatatgcaaggtaagttttggtgctaataggccaatgttttcataatgtaaa 120
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Sbjct: 2809 agcaattcaatatgcaaggtaagttttggtgctaataggccaatgttttcataatgtaaa 2868
integrin alpha 2 56 A I Q Y A R~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Query: 121 acattatatttatgtaataaatatgaaaaagtaaggaaaagacaaagaaaaataatatac 180
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Sbjct: 2869 acattatatttatgtaataaatatgaaaaagtaaggaaaagacaaagaaaaataatatac 2928
integrin alpha 2 2869 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Query: 181 ctggtacctaatttaaatcagaactaataaagnnnnnnncatcagagcattctatgtctt 240
|||||||||||||||||||||||||||||||| ||||||||||||| |||||||
Sbjct: 2929 ctggtacctaatttaaatcagaactaataaagaaaaaaacatcagagcattcaatgtctt 2988
integrin alpha 2 2929 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Query: 241 gaatactttgagaaggcagctgggaaagttaaa 273
||||||||| |||||||||||||||||||||||
Sbjct: 2989 gaatactttaagaaggcagctgggaaagttaaa 3021
integrin alpha 2 2989 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
The antisense primer also matches to a fragment (5616-5668) on the same gene but different strand. This match can be ignored because there are some mismatches on the primer's 3'end and is not on the same strand as the forward primer. They wont cause non-specific amplification.
||||||||||||||||||||||||||||| ||||||||| ||||||||||||
Sbjct: 5668 atcagagcattctatgtcttgaatactttaagaaggcagtcgggaaagttaaa 5616
So how can I improve my query and find even those non-specific matches?
To get other potential non-specific matches, just lax your parameters such as Min Perfect Match and Min Good Match, but I think using the default parameters is good enough.
I am not shure if I really got you. With the blat-search I get only one result at all. What should a change of the parameters such as Min Perfect Match/ Min Good Match make?
The parameters tell the program to just return targets that must have for example 15 bases matched to one of your primer. If you lower the number to 10, you will probably get lots of matches.
Sorry, how do you get lots of matches with BLAT or where can I change the parameters such as Min Perfect Match/ Min Good Match in BLAST?