You cannot simply fix error bars; you need to determine the sources of variation and, if necessary, repeat the experiments. Are you referring to technical or biological replicates? Major differences between technical replicates could be due to pipetting errors, low primer efficiency, etc. For biological replicates, some variation is expected, but ensure you use a standardized experimental setup to minimize deviations in the results.

Hello,

I wouldn't recommend comparing transcriptome to qRTPCR since they are completely different techniques. The problem essentially comes from the chosen genes (both reference and target) and primers. There are some tools you can use you can use to help you choose genes after transcriptome experience: https://bmcgenomics.biomedcentral.com/articles/10.1186/s12864-024-10511y#:~:text=Real%2Dtime%20quantitative%20PCR%20(RT,conditions%20present%20in%20the%20transcriptome . What is suggested is to use a reference gene that has both a high expression and a stable expression in your biological conditions (sometimes even RPS, ACT or GAPDH are not good candidates). Target gene should have a big variability and target primers should target exons....

I would still trust transcriptome data ;)

Sorry link is not working. I want to know, yes both techniques are different but still they give me the same feedback of some genes, as I found in some research paper, they mix between qpcr and transcriptome analysis in one graph by using log2 fold change