Hi, now I'm preparing library for SMART-seq2 using low input total RNA (~10 ng)

The protocol i used is based on this paper: Picelli, S., Faridani, O. R., Bjorklund, A. K., Winberg, G., Sagasser, S. & Sandberg, R. (2014). Full-length RNA-seq from single cells using Smart-seq2. Nat Protoc, 9(1), 171-181. doi: 10.1038/nprot.2014.006

My experiment design involves control vs knock-down of GOI samples.

Prior to the NGS library preparation, I checked my expression of knock-down gene using original cDNA (without pre-amplification), and the gene expression level is reduced as expected.

Based on the protocol, I did pre-amplification using KAPA HiFi HotStart Library Amplification Kit.

Then I checked the qPCR again to confirm the differential expression of my knock-down GOI with the control.

However, some replicates do not get the same result in which my GOI expression level is not reduced, and some of them even higher than the control.

I'm confused if the pre-amplification method has introduced some bias to the cDNA genes, causing lack of preservation of differential expression after preamplification.

Does qPCR is invalid to measure the differential expression after preamplification?

I used ERCC spike-in, and does ERCC normalisation will solve this different expression level?

Thank you