Hi,
I am pasting here the methods from a Standard Operating Procedure (SOP) that I wrote many years ago with a colleague for designing primers (or primer + probe). This protocol has literally never failed us, i.e., each set of primers that have ever been developed by following this SOP has functioned.
5. PROTOCOL
5.1
Obtain the start and end genomic co-ordinates for a region of interest
5.2 Obtain the DNA sequence for the region of interest using the NCBI’s ‘Map Viewer’ tool.
- i. Navigate to http://www.ncbi.nlm.nih.gov/mapview/ (now https://www.ncbi.nlm.nih.gov/genome/gdv/ )
- ii. Select the species and assembly, and then enter genomic co-ordinates in 'Search in genome' text box, e.g.,
chr8:100000-100100 - iii. On the right, click on ‘Download’ -> 'Download FASTA' -> 'FASTA (Visible Range)'
- iv. Save / Copy the DNA sequence that appears
5.3 Check that the DNA sequence of interest has little or no homology with other regions in the human genome.
Only proceed from this step if there are no other homologous sequences, as these can misrepresent the result of PCR.
- i. Navigate to http://blast.ncbi.nlm.nih.gov/ and select ‘nucleotide blast’
- ii. Paste the DNA sequence into the text box, select ‘Human genomic + transcript’ database, and click ‘BLAST’ (The parameters will automatically adjust if the sequence that you’ve entered is short)
- iii. On the results page, the alignment score is shown in coloured boxes (black across to red). If the sequence is a true sequence from the human genome, then there will always be a high alignment score (red line) with a query coverage of 100% that matches the true sequence. Any homologous regions elsewhere in the genome will generally have a lower alignment score (and different colour) and lesser query coverage.
5.4 Check the level of SNP coverage and other short variations in the DNA sequence of interest.
This step is not critical; however, the fewer SNPs and short variants, the more reliable the PCR result will be.
- i. Navigate to http://blast.ncbi.nlm.nih.gov/Blast.cgi?PROGRAM=blastn&BLAST_SPEC=SNP&BLAST_PROGRAMS=megaBlast&PAGE_TYPE=BlastSearch
- ii. Paste the DNA sequence into the text box, select ‘Homo sapiens’ from the drop-down list, and click ‘BLAST’ (The parameters will automatically adjust if the sequence that you’ve entered is short)
- iii. A graphical display of SNP and short variant hits will display
Nota bene 1: NCBI's SNP BLAST seems to have been 'retired'. One can also visually search for short variants via the UCSC Genome Browser: https://genome.ucsc.edu/
5.5 Design the primer and probe sequences using Primer3
- i. Navigate to http://primer3.wi.mit.edu/ ( now https://primer3.ut.ee/ )
- ii. Paste the DNA sequence into the text box and ensure that ‘Pick hybridization probe’ check-box is checked
iii. Configure the search using the following parameters:
a. ‘Primer Tm’, 60-66
b. ‘Primer GC%’, 40-60
c. ‘Product Size Ranges’, 50-120
Although these are not the ideal ranges, Primer3 has a tendency to overestimate values, which is why we also overestimate
- iv. Click on ‘Pick Primers’ (also chooses the probe). A new page will appear showing again the DNA sequence that was input; however, the automatically-generated forward (>) and reverse (<) primer and probe (^) will also appear (all in 5’-3’ order)
Nota bene 2: Primer and probe sequences need to follow general guidelines [see Additional Notes, at end]
Nota bene 3: ignore the parts relating to the probe if you are just developing standard primers
5.6 Keeping Primer3 open, check the primer and probe sequences in Primer Express
- i. In Primer Express, go to ‘Tools’ and then ‘Primer Probe Test Tool...’
- ii. Copy the primer and probe sequences from Primer3 and paste into the text boxes in the test tool
- iii. If the sequences are thermodynamically sound, then the ‘Tm’ and ‘%GC’ will appear green. If not, then some bases need to be either removed or added from/to either end
- iv. Check the primers for predicted secondary structures (beneath)
Nota bene 4: you probably do not need step 5.6 and, irrespective, Primer Express is a commercial program.
5.7 Repeat steps 5.5 and 5.6 until a suitable set of primers and probe is designed, changing the parameters at different steps, if necessary.
5.8 Check which region of the human genome is amplified using in silico PCR
- i. Navigate to http://genome.ucsc.edu/, select ‘In-Silico PCR’, and ensure that the correct reference sequence is selected in ‘Assembly’
- ii. Copy the primer sequences and paste into the ‘Forward Primer’ and ‘Reverse Primer’ text boxes
- iii. Click on ‘submit’. A new page will appear giving details of the hypothetically-amplified regions. Ideally, there should only be one region. If not, then be aware that the probe would add some further level of specificity; however, if not happy, then consider restarting from step 5.1!
6. ADDITIONAL NOTES
General guidelines for primer and probe sequences:
Primers
- i. Optimum amplicon length, 50-120 nucleotides
- ii. Optimum primer length, 20 nucleotides
- iii. Optimal Tm, 59 Celsius
- iv. 3’ end should have no more than 2 Gs or Cs in the final 5 nucleotides
- v. Keep homopolymers (repeating nucleotides) to a minimum; if present, there should be no more than 4 Gs together
FAM-labelled, MGB probe
- i. Optimum probe length, 13-25 nucleotides
- ii. Optimum Tm, 68-70 Celsius
- iii. 5’ last nucleotide cannot be a G; neither can the penultimate (second-from-last) nucleotide
- iv. 3’ end should have few Gs; specifically avoid GGG and GGAG
v. Keep homopolymers (repeating nucleotides) to a minimum, specifically:
a. No more than 4 Gs together
b. No more than 6 As together
c. No more than 2 CC dinucleotides in the middle of the probe
Credits:
- Kevin, Karen, and Jacqui
- University of Leicester
Kevin
Note that Primer-BLAST will perform many of these steps automatically (and has similarly sensible default values) and is quite nice for designing qPCR primers that span a junction.