Hi all,
I am trying to verify the fold change of 18 genes (early stationary phase E. coli, control vs. treated, reference used is 16s rRNA) observed in RNAseq results by RT-qPCR, but was unable to produce similar fold change data in qPCR. For example, my pspG gene showed 500-1,000 fold up-regulation in treated cells in RNAseq, but was only found to have 8-16 fold up-regulation in qPCR. Most other genes with smaller fold changes (e.g. 5-30 fold) in RNAseq was shown to have no up-regulation (sometimes even down-regulation) in qPCR.
Troubleshoot done so far:
1) RNAseq error?
No, all samples were sent in triplicate and produced consistent results of fold change.
2) qPCR error? -
Total RNA quality: confirmed OK by Nanodrop and gel electrophoresis
- Genomic DNA contamination in total RNA: confirmed No by gel electrophoresis
- Primer problem (because self-designed): confirmed OK by melt curve (single peak) and gel electrophoresis (single band)
- cDNA problem: confirmed OK by running cDNA with 16s rRNA primers (primer sequence from published papers), and prepared 3 batches of cDNA all generated similar fold change results
- qPCR master mix problem: tried 2 brands (Promega, BioRad), generated similar fold change results
- qPCR machine problem: tried 2 machines, generated similar fold change results
Any suggestions on what other factors I can check for potential qPCR errors? Or if anyone has experienced a similar “lack of correlation between RNAseq and qPCR data” problem? I’ve read that people experiencing “poor correlation for genes with low expression level”, but is 5-30 fold considered very small?
Kindly let me know if any additional infor would be helpful, and thank you in advance for any suggestions/advices!
these links may be useful
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC1544370/
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5218489/
https://support.bioconductor.org/p/95949/
https://bitesizebio.com/2345/troubleshooting-rna-isolation/
regards
Hello Wenqian:
It seems that this is because the number of ref. genes is too short, you should have at least 3 ref. genes and the geometric mean as the reference method.
all other quality controls seems to be really good!!
I advice you to find other 3 ref genes since 16srRNA is always overexpressed in contrast with all of the other genes, which make the control really variable.
both methods are somewhat different procedure. Sensitivity of Rnaseq is more than qPCR . in qPCR what you are using probe or dye it also can affect.and on your RNA extraction. if result of both are same like upregulation or downregulation. then only sensitivity issue of both procedure.
- Badges
- Science method