RT-qPCR vs qPCR: They're Not the Same Assay

RT-qPCR and qPCR are not interchangeable terms, even though half the papers on PubMed use them that way. qPCR (quantitative PCR) amplifies and quantifies a DNA template in real time. RT-qPCR (reverse transcription qPCR) adds an upstream step: converting RNA into cDNA before the qPCR reaction runs. If your starting material is RNA — total RNA from cells, tissue, or sorted populations — you are doing RT-qPCR. If you're quantifying a DNA target directly — viral DNA load, transgene copy number, ChIP-enriched fragments — that's qPCR. The reverse transcription step is the entire difference, and it introduces its own set of variables, controls, and failure modes.

Why does this matter beyond nomenclature pedantry? Because the RT step is often the largest source of technical variability in the whole workflow. Two identical RNA samples reverse-transcribed in separate reactions can easily give you Ct values that differ by 1–2 cycles, which is a 2–4 fold difference in apparent expression. If you don't understand which part of your protocol is RT-qPCR and which is qPCR, you can't troubleshoot effectively, you'll design the wrong controls, and your reviewers will (rightly) push back.

What Happens in Each Assay

In a qPCR experiment, you load a DNA template — genomic DNA, plasmid, ChIP DNA, bisulfite-converted DNA — into a reaction containing DNA polymerase (usually a hot-start Taq), primers, dNTPs, and a detection chemistry (SYBR Green or a TaqMan probe). The thermocycler runs denaturation/annealing/extension cycles while reading fluorescence in real time. You get a Ct (or Cq) value that's proportional to the log of your starting template amount. That's it. One enzyme, one process.

In RT-qPCR, you first convert your RNA into complementary DNA (cDNA) using a reverse transcriptase — common ones include SuperScript IV, Maxima H Minus, or the RT packaged in Luna or PowerUp kits. This step requires its own primer (oligo-dT, random hexamers, or gene-specific primers), runs at a specific temperature (typically 42–55°C depending on the enzyme), and takes 10–30 minutes. The resulting cDNA then becomes the template for a standard qPCR reaction.

You can run this as a two-step or one-step protocol:

• Two-step RT-qPCR: Reverse transcription happens in a separate tube. You generate a cDNA pool, then aliquot it into multiple qPCR reactions for different targets. This is what most gene expression studies use because you can assay many genes from one RT reaction.
• One-step RT-qPCR: RT and qPCR happen sequentially in the same tube — the thermocycler runs an RT incubation phase before starting PCR cycles. Common for viral diagnostics (yes, those COVID tests were one-step RT-qPCR) and high-throughput screens where you're measuring one or two targets per sample.

Each approach has tradeoffs. Two-step gives you flexibility and a cDNA archive. One-step reduces pipetting (fewer contamination opportunities) but locks you into one target per well and uses more RNA per target.

Why the RT Step Is Where Things Go Wrong

The qPCR portion of the assay is, frankly, pretty robust once your primers are validated. Modern master mixes on a QuantStudio 5 or CFX96 are forgiving. The reverse transcription step is where variability hides.

RT efficiency is not 100%, and it's not consistent. Different transcripts reverse-transcribe with different efficiencies depending on secondary structure, GC content, and transcript length. A highly structured 5' UTR on your gene of interest can suppress cDNA yield for that specific transcript, making it look underexpressed relative to a reference gene like ACTB that reverse-transcribes efficiently. This is a real and underappreciated confounder.

Priming strategy matters. Oligo-dT primers select for polyadenylated mRNA and give you 3' bias — if your qPCR amplicon is near the 5' end of a long transcript, you may get incomplete cDNA and artificially high Ct values. Random hexamers prime everywhere (including rRNA, which is ~80% of your total RNA) but give more uniform coverage. Many two-step kits use a blend of both, which is a reasonable default. Gene-specific primers give the highest specificity but only produce cDNA for that one target, so you lose the ability to go back and measure other genes.

Genomic DNA contamination is an RT-qPCR problem, not a qPCR problem. If you're amplifying from RNA-derived cDNA and your primers sit within a single exon, genomic DNA in your RNA prep will amplify right alongside the cDNA. This is why you need:

1. DNase treatment — either on-column during RNA extraction or in-solution before RT. I prefer on-column (e.g., Qiagen RNeasy with the DNase I add-on) because it doesn't require a cleanup step that loses RNA.
2. Intron-spanning primers — design your qPCR amplicon to cross an exon-exon junction so that genomic DNA either produces no product (if the intron is too large) or a product of a clearly different size.
3. NRT controls — a no-reverse-transcriptase control uses the same RNA but omits the RT enzyme. Any Ct you see in the NRT is genomic DNA signal. If your NRT Ct is >5 cycles higher than your +RT Ct (meaning <3% contamination), most people consider that acceptable. If it's within 3 cycles, you have a gDNA problem that needs fixing before you trust any expression data.

None of these controls are relevant in a straight qPCR assay where your template is already DNA.

Controls and Normalization Differ Too

Both assays use a no-template control (NTC) — water instead of template — to catch contamination or primer dimers. Ideally the NTC shows no amplification. If you see a Ct of 38–40 in a SYBR Green NTC, check the melt curve: a single sharp peak at your amplicon's Tm means low-level contamination; a broad peak at 72–78°C is primer dimer and usually ignorable as long as your samples are coming in below Ct 32 or so.

For RT-qPCR specifically, you need the NRT control described above. You also need to think carefully about reference gene selection. Normalizing GOI expression to GAPDH works fine in many cell culture experiments, but GAPDH is notoriously variable across metabolic states, hypoxia, and between tissues. For cross-tissue comparisons, something like HPRT1, TBP, or B2M often performs better. Tools like geNorm (Vandesompele et al., 2002) and NormFinder (Andersen et al., 2004) can rank candidate reference genes by expression stability across your sample set. Using two or three validated reference genes and taking the geometric mean of their Ct values is the current best practice.

In DNA-based qPCR — say, copy number variation analysis — normalization is typically to a single-copy reference locus like RNase P or TERT rather than a "housekeeping gene," and the math (2^−ΔΔCt or a standard curve approach) is applied to DNA quantity directly.

The Math Is the Same, the Inputs Aren't

Whether it's RT-qPCR or qPCR, the quantification math is identical. The Livak method (Livak and Schmittgen, 2001) gives you 2^−ΔΔCt fold change assuming all primer pairs have ~100% efficiency. The Pfaffl method (Pfaffl, 2001) adjusts for unequal efficiencies between target and reference. In both cases, you calculate ΔCt = Ct_GOI − Ct_REF within each sample, then ΔΔCt = ΔCt_treated − ΔCt_control across conditions.

The key difference is what those Ct values represent. In RT-qPCR, a Ct of 25 reflects the abundance of a given mRNA transcript (filtered through RNA extraction efficiency, RT efficiency, and PCR efficiency). In qPCR, a Ct of 25 reflects the abundance of a DNA target. Conflating them leads to nonsensical comparisons — you cannot compare a Ct from a ChIP-qPCR experiment to a Ct from an RT-qPCR experiment and draw any meaningful conclusion.

Statistical analysis should be performed on ΔCt values, not on fold changes. Fold changes (2^−ΔΔCt) are ratios and are not normally distributed; ΔCt values are log-transformed and generally suitable for parametric tests. Use an unpaired t-test for two groups or one-way ANOVA for multiple groups, applied to ΔCt values. Report fold changes in figures for biological interpretability, but run your stats on the ΔCt.

Getting the Terminology Right in Your Manuscripts

It's a small thing, but using the correct term signals that you understand your own assay. If you extracted RNA and made cDNA, write "RT-qPCR" in your methods. If you quantified genomic DNA or plasmid, write "qPCR." The MIQE guidelines (Bustin et al., 2009) — which most journals now reference in their author instructions — are explicit about this distinction and about the minimum information you should report (primer sequences, reaction conditions, efficiency values, reference genes used, and how normalization was performed).

One more thing that trips people up: "real-time PCR" and "quantitative PCR" mean the same thing (qPCR). "RT-PCR" without the "q" technically means reverse transcription PCR analyzed on a gel — an endpoint assay, not real-time. In practice, people are sloppy about this, but if a reviewer circles "RT-PCR" in your manuscript and writes "do you mean RT-qPCR?", they're not being pedantic. They're asking whether you quantified fluorescence in real time or just ran a gel. Answer clearly.

If you're running RT-qPCR and want the ΔΔCt analysis, efficiency checks, reference gene stability, and NRT flagging handled without building another spreadsheet, upload your data to VoilaPCR — it runs these checks automatically and catches the issues that matter.