Hello All,
I'm in the qPCR validation step of RNAseq gene expression and my qPCR calculated logFC is showing as upregulated for all my test genes, including predicted downregulated genes. This is in citrus plant samples that have been infected with a viral pathogen. The treatment is virus infected and the control was mock infected.
The bioinformatics was performed with BioBam OmicsBox, which everything I did for the RNAseq was "standard" settings. I verified with MDS that the samples have distinct clustering and I looked at the raw counts, which also follow the expected pattern of up- and downregulated genes. I have performed bioinformatics within omicsbox previously and had the expected qPCR results on another citrus gene expression study (down matched down, up matched up), so I wouldn't think that I performed the bioinformatic pipeline incorrectly or any differently for this study.
I have 3 reference genes that have all been validated, and I have tried them all with my samples and received the same result, downreg genes show strong upregulation.
For the samples used in qPCR, I DNase treated total RNA with a column cleanup to create cDNA for the qPCR. I also verified with a negative control cDNA (no RT enzyme) that I have no qPCR amplification, so no genomic DNA carry over after DNase.
Currently I'm trying to figure out if my primers are hitting on any other targets. The one large variable I have is that my reference genome is new and to my knowledge, hasnt had an RNAseq experiment performed with it yet. Currently I'm building a local database in OmicsBox with this reference genome and my viral genome to verify if the primers are hitting on more than one target.
Any other suggestions on how to troubleshoot this problem would be greatly appreciated. Thanks in advance.
This is a frustrating situation that many researchers run into! The fact that all your "downregulated" genes are showing up as upregulated in qPCR suggests there's likely a systematic issue rather than random primer problems.
Here are some things to check:
Sample mix-ups are more common than we'd like to admit. Double-check that your qPCR samples actually correspond to the same RNA samples used for RNAseq. It's easy to accidentally swap treatment and control labels somewhere along the way, especially if you're working with multiple timepoints or conditions.
Your primer specificity check is smart - that new reference genome could definitely be the culprit. If your primers are hitting multiple targets or the wrong targets entirely, that would explain the consistent pattern you're seeing.
Reference gene stability might be an issue even if they've been "validated" before. Viral infection can mess with housekeeping gene expression in unexpected ways. Try plotting your reference gene Ct values across all your samples - they should be fairly consistent. If they're all over the place, that could throw off your calculations.
RNA degradation is another possibility. Even if your samples looked good initially, degradation during storage or processing could skew results. Check if you still have some RNA left to run on a gel or Bioanalyzer.
Different RNA extraction batches between your RNAseq and qPCR samples could also cause issues, especially if the extraction efficiency varied between treatments.
Since you've done this successfully before, I'd lean toward either a sample labelling mix-up or primer specificity issues with the new genome. The primer database check should help clarify that second possibility. Hope that helps.
- Badges
- Science method