Our lab has an assay to detect fusion transcripts that very frequently results in "false positives." Often our duplicates show mixed results (one positive and one negative), always at a high C(t), and when re-run they most often turn out negative, but occasionally mixed results again. Our RT-PCR uses random hexamer primers, followed by Q-PCR with transcript-specific primers.
When these noisy samples are run on a gel, many bands appear of various sizes. A number of these bands are LARGER than the target amplicon. (Target transcript is 92bp, bands in noisy samples span ~55-205bp). Suggests non-specific primer binding, right? However, Primer BLAST shows no possible non-target binding sites for this primer set, and our annealing temperature is higher than recommended to discourage non-specific binding.
These are actual amplicons, and appear not to be probe degradation or other noise from the qPCR.
We are trying to eliminate this noise but can't determine the source. How reliable is Primer-BLAST for identifying alternative binding sites for a primer set? Is there a way RNA degradation could result in amplicons that are larger than the target sequence? (We do check RNA quality by checking for 18S and 28S bands). Any suggestions for trouble shooting this?
Hi..
It sounds like you're having a difficult time identifying the source of the "false positives" in your assay. The fact that the amplicons from the "noisy" samples are larger than the target transcript suggests that they may be resulting from non-specific binding of the primers. However, you mention that PrimerBLAST shows no possible non-target binding sites for the primer set, and that the annealing temperature is higher than recommended to discourage non-specific binding, which makes it less likely that non-specific binding is the cause of the problem.
Primer-BLAST is a reliable tool for identifying alternative binding sites for a primer set, but it is not perfect. It is possible that the primer set may be binding to a site that is not present in the database used by Primer-BLAST, or that the algorithm is not taking into account some aspect of primer binding that is important in your specific case.
RNA degradation could result in amplicons that are larger than the target sequence if the degradation products contain sequences that are complementary to the primer set. This could happen if the primers are binding to partially degraded RNA molecules, or if the RNA is being degraded in such a way that the primer binding site is preserved. One way to test for this possibility is to run the PCR reaction with a higher annealing temperature, which should make it more difficult for the primers to bind to degraded RNA molecules.
It's also possible that non-specific binding is happening but with non-target sequence that is not in the primerblast search database , or that you are amplifying a rare or previously unknown isoform, or that contamination of your samples or reagents is causing the problem. Another way to try and pinpoint the problem is using a technique like Multiplex PCR, where you are using several primer sets that target different regions of the same transcript to see which primer set causes the false positives. It may also be helpful to analyze the sequences of the amplicons generated from the "noisy" samples to see if they match any known sequences, which could help to identify the source of the problem.
In general, it is important to maintain strict experimental controls and keep detailed records of the conditions used in the assay to help identify the source of the problem. It may also be useful to consult with other experts in the field or a colleague that has experience with similar problems.
Regards
Thank you for the information. I don't know anyone else with the same problem! That's why I'm asking here.
We do run positive and negative controls in these assays, and have no problems with them.
In the noisy samples there appear to be many bands that are similar in size, but the banding pattern is inconsistent between samples (each band does not appear in every sample). I would imagine there should be only one or two alternative binding sites, not 7-10, especially since primerBLAST can't find any, is that correct?
I think it could be possible that there are multiple isoforms that are showing up. RNA degradation is also possible, I think, despite the 18S and 28S quality checks. I'd like to sequence a couple of the stronger bands, but have logistical barriers that make this a course of last resort. I'm interested in your multiplex idea.
Hello
One of the methods that can help this issue is to optimize the temperature.
- Badges
- Science method