Question: Primer design using Bioinformatics tools.
gravatar for siddharth.avadhanam
4.7 years ago by
siddharth.avadhanam30 wrote:

I am currently working on a project designing a primer template for detecting any pathogenic species of leptospira in a serum sample, but I seem to have hit a roadblock. While the ncbi primer design tool has proven exceedingly useful, I am unable to figure out how to extract a conserved primer (both forward and reverse) that can be used for two different strains of borgpetersenii and three different strains of interoganns. I used mauve to get a comparative genomics profile of the five different strains, and then ran a highly conserved region through the ncbi pcr design tool. Wile it gives me results, these results are good for the particular species that I used. So how could i go about designing a primer that can be used for all these strains, if that is at all possible ? Any help or clarity would be greatly appreciated. Thanks in advance 

blast genome • 5.1k views
ADD COMMENTlink written 4.7 years ago by siddharth.avadhanam30

You can use ProSig ( to automatically design PCR primers that target conserved regions of multiple bacterial genomes. The basic idea is to first enumerate a large number of valid PCR primer pairs and then throw away all of the pairs that don't produce an amplicon in all of your target genomes. This approach does not require a multiple sequence alignment. See for a description of the algorithm.

ADD REPLYlink written 4.7 years ago by JasonGans100

The link to ProSig no longer works. The publication has only been cited 8 times since 2012.

ADD REPLYlink modified 8 months ago • written 8 months ago by Kevin Blighe41k
gravatar for Gjain
4.7 years ago by
Göttingen, Germany
Gjain5.3k wrote:

I hope this might be useful to you:

ADD COMMENTlink written 4.7 years ago by Gjain5.3k

I think what he needs are primers that would anneal to different templates with comparable fidelity. Primer3 to my knowledge does not provide such options. My thought is to design several pairs of primers (as few as possible) to cover all the genomes of interest. There should also be some level of redundancy so that you can normalise against each other?

ADD REPLYlink written 4.7 years ago by logust7920

aah thanks ... got it ... but why not design the primers and barcode them ?

ADD REPLYlink modified 4.7 years ago • written 4.7 years ago by Gjain5.3k

Thanks for the replies. So this necessarily has to be a trial and error processes? Maybe I could keep blasting a primer I get from one species against all the leptospira strains on the ncbi database ? Are there any other tools that I could use? Also, ( and bear with me here, I am very new to this field ) what do you mean by normalise them against each other ? 

ADD REPLYlink written 4.7 years ago by siddharth.avadhanam30

This is what I think. Say you want to look at gene x in A..E different species. I would do alignment of x from A..E, and look for several highly conserved regions.

These regions have to cover all the variants of x from A..E. This is the fun part. Because you need two primers for PCR, and the conserved regions might not be big enough for you to fit two primers in it (especially when you have many species to be taken care of). And if you are looking at Real-time PCR, then there is also a template size limit. The way to turn around this is:

primer F1 <- perfectly conserved in A,B,C; primer R1 <- perfectly conserved in A,B,D; then product 1 is confident in A, B.

primer F2 and R2 -> product 2 (as above) is confident in C,D,E

Then using primers F1/R1 and F2/R2 should be able to cover all the genomes. However, since F1/R1 and F2/R2 are different primers, they might have different 'powers' in amplification. If you want to quantify the outcome (say how much A..E in the sample?), this could confound your results. So you might also need F3/R3, which is confident in B, C (overlap with product 1 and 2) to 'normalise' the results you get from product 1 and 2, to get rid of the confounding factor.

Of course I'm also new to the field, so I'm probably wrong, or not knowing anything already exists to deal with such a problem?

ADD REPLYlink written 4.7 years ago by logust7920
Please log in to add an answer.


Use of this site constitutes acceptance of our User Agreement and Privacy Policy.
Powered by Biostar version 2.3.0
Traffic: 1200 users visited in the last hour