ERCC spike-in analysis

Question

[anilkumar.vijjamarri]

Hi, I recently performed RNA-seq experiment with ERCC spike-ins. I added the recommended amounts of mix1 to WT and mix2 to mutant total RNA before library preparation. I calculated the observed ratios (log2 RPKM >=1) vs expected ratios and found that the R2 value about 0.90 but the slope is ~0.7. This indicates that the observed ratios are some what off to expected ratios. I do not understand whether I can use the slope from equation to normalize the reads for all the genes?

I also plotted the log2 RPKM values against their expected concentrations and also found good correlation (0.95) but the slope is 0.77.

Can any explain whats happening and whether I can use this for normalization of the genes?

Thanks, Anil

Answer 1

[JC]

The use of ERCC spike-ins is intended to give a reference for normalization (How does edgeR do ERCC spike-in normalization?) however, these added sequences are not recommended https://pubmed.ncbi.nlm.nih.gov/25150836/