aligning sequences for primer design later

Question

[nouranatef.khalil]

Hi all,

I am a Microbiologist, and I'm very new to using any bioinformatics tool and still struggling in understanding the very basic concepts of bioinformatics in general....I hope you would bear up with my very naive questions, as I'm really really lost..

I want to design a primer for my bacterial enzyme, so from what I know is that I have to make multiple alignment for the DNA sequences I have first.. .

I am using CLC genome workbench as I find it user friendly. I add about 8 different sequences of this enzyme from 8 different bacterial strains and then made the alignment, but doing this I found that there is almost no similarity and the percentage of conserved parts is very little. I understand this would be a hindrance to design a good primer, right?

So what shall I do now...should I choose other organisms than those and start all over or what?

Answer 1

[Andrzej Zielezinski]

Highly variable sequences are hard nuts to crack, but I would try PrimerDesign-M software published recently in Bioinformatics (see the paper). This tool designs primers based on multiple-alignment of highly variable genomic sequences.

Answer 2

[Mohamed]

You need to specify your aim from this designing. Are you going to do clone for expression, in this case, you need from the starting methionine to the last stop codon. Or, if you want to only isolate the conserved region(s). Also, you should mention if you aligned DNA/mRNA or Protein seq. I got the feeling that you might doing something wrong. Mostly aligning similar sequences will give you conserved region all over the seq. Also, remember that mRNA or CDS seq will not help you if you wish to clone for expression, you need to find where is the DNA/mRNA seq that correspond to the starting aa. The same for the last stop codon. Another hint is to check if you use CLC GW properly or not, is by copying and pasting one seq and use it in your alignment.

[nouranatef.khalil]

I'm designing a primer for the identification of the seq of the enzyme in my organism..

First I logged to NCBI and chose the nucleotide database, then in the search I wrote the name of my enzyme (ex:levansucrase) and chose Bacteria and refseq in the filters..

I chose completely random and different bacteria and copied the CDS of the enzyme I'm looking for, I chose the FASTA format and made sure that the sequences start with ATG.. and then made the alignment..

But then I was told, that It's better to specify my search to different species of my organism (in my case it's Bacillus), so instead of looking among completely different bacteria I specify my search to the bacteria I'm interested in to have better alignment results..

Am I on the right track?

Thanks a lot for your help :)

[Mohamed]

yes. It is better to have more sequences and from the same species to get conserved regions. Actually, I prefer that you do design two set of primers. One based on your own very similar species, another set based on downloading first so many similar sequences (make sure that at least they are about the same length, at least), make alignment and then design the primers. For such task, I prefer not to use primer3 but rather do it manually like oligocalc. Also, do not be afraid from dimer primers and/or the length of the primers (actually sometimes I use one of the primers with 35 bases and it works with another primer that is 21 bases (imagine the difference in Tm between them and it works).