The workflow was just scripted according to what was already routinely performed in the laboratory by wet-lab biologist with a decade of experience performing PCRs and sequencing.
- First SNV positions are verified (manually) to be hg19. If not, they are lifted over from hg18.
- Flanking regions of the SNV – 300 bp on either side are selected from hg19 reference.
- These are put through primer3 in batch (and redesigned the second time with relaxed parameters, if selected).
- The primer pairs are then in-silico tested to see if they are unique and flagged if not.
- Then the SNV is checked to see if paired-end sequencing (of the specified read length) will read the position from both the forward and reverse directions.
- The amplicons (in lowercase) passing the filters are shown (so you can manually check for low complexity/repeats etc) and their respective primers are identified at the ends (UPPERCASE).
- A primer order file is automatically generated as is the AmpliconManifest file for MiSeq for each case. Please see the end of the output for the exact path on the server to download your files.
This is a list of features that we have implemented and published but we have not ported the version to this web version yet. Please email email@example.com if you think they would be useful and we will put it on our priority list according to how much interest is received.
Yet to be implemented in this web version
- Input of flanking DNA sequence with unique patient-specific DNA including personal SNPs (there might be privacy issues with this – there we have not implemented it on the web version). This is so primer3 can avoid designing primers which fail to bind due to private patient-specific SNPs or design patient-specific primers which always bind.
- Output of HTML format of the sequence and the position of the primers relative to the SNV of interest. This would be to manually inspect the amplicon to see if it is complex or repetitive. If it is repetitive – although the primer regions may be complex enough to be specific, during alignment phase of the sequenced amplicon (ie fastq output from the MiSeq), loss of information at that stage, could result.
- Automated addition of either Illumina or Fluidigm primers so that ordering can be done right away, streamlining the process.
- There have also been suggestions to design primers over a range rather than a specific position. However, since we have not done that, the process remains unvalidated. Although it could easily be implemented.
This has been a side project of a wet-lab biologist for other wet lab biologists who routinely design primers for the deep-sequencing of SNPs. We have received emails scientists from all over the world – from the UK, to France, to Australia, telling me they have used the web work flow.
Most laboratories would now have their own bioinformatics staff who could easily write such a workflow for in-house use. However, this was developed in 2010-2011, where such a set up was not common. In the meantime, we are pleased to enable scientists by making this validated tool readily available. They can rest assured that the process has been validated and reference the various publications in which the workflow has been deployed previously.