qPCR vs RT-qPCR: The Key Differences and When to Use Each

The qPCR vs RT-qPCR question usually gets answered in one line: qPCR is for DNA, RT-qPCR is for RNA. That is correct, and it is where most explanations stop. It is also not the part that costs people data.

The part that costs people data is what the extra step actually does to your numbers. Adding reverse transcription to the front of a qPCR reaction does not simply extend the workflow. It introduces an enzymatic conversion whose yield is not fixed, not necessarily proportional to input, and not the same for every transcript in the tube. Published work has found reverse transcription yield varying by up to 100-fold depending on the priming strategy, the enzyme, the amount of starting RNA, and the sequence being detected.

That is the real content of the comparison. Below: the nomenclature (which is genuinely confusing and is confused in the literature), how each assay works, the controls each one demands, the validation criteria, and how to decide which one your target requires.

qPCR vs RT-qPCR: The Short Version

qPCR — quantitative real-time PCR — amplifies a DNA template and measures product accumulation cycle by cycle through fluorescence. The cycle at which the signal crosses a threshold, the quantification cycle (Cq), is inversely proportional to the log of starting template. Input is DNA. That means genomic DNA, plasmid DNA, cDNA, or the genome of a DNA virus or a bacterium.

RT-qPCR — reverse transcription quantitative PCR — starts from RNA. A reverse transcriptase converts the RNA into complementary DNA, and that cDNA becomes the template for an otherwise ordinary qPCR. Input is RNA: messenger RNA, viral genomic RNA, ribosomal RNA, microRNA.

The amplification and detection chemistry is identical. The difference is entirely upstream. Everything else in this article follows from that one added enzymatic step.

Fix the Nomenclature First

This is worth two minutes, because the terminology is used inconsistently even in peer-reviewed papers and it causes real confusion when reading protocols.

The MIQE guidelines — Minimum Information for Publication of Quantitative Real-Time PCR Experiments, published by Bustin and colleagues in Clinical Chemistry in 2009 — set the convention that the field now uses:

TermWhat it meansWhat it does not meanPCREndpoint polymerase chain reaction—RT-PCRReverse transcription PCR, legacy endpoint formatIt does not mean "real-time PCR"qPCRQuantitative real-time PCR, DNA template—RT-qPCRReverse transcription qPCR, RNA template—

The single most common error is reading the RT in RT-PCR as "real-time." It stands for reverse transcription. The q is what signals real-time methodology. So it is qPCR for DNA and RT-qPCR for RNA, and RT-PCR refers to the older endpoint reverse transcription method that resolves products on a gel.

Two related conventions from the same guidelines: report Cq (quantification cycle) rather than Ct, Cp, or TOP — those were instrument-vendor terms for the same value — and refer to hydrolysis probes rather than the trademarked TaqMan, and to reference genes rather than housekeeping genes.

MIQE was revised as MIQE 2.0, published in Clinical Chemistry in 2025. The original has accumulated over 17,000 citations and has shaped journal editorial policy and ISO standards for molecular diagnostics. The revision replaces the old essential/desirable tiering with a single unified checklist organised into five sections: reagent preparation, sample preparation, reverse transcription, qPCR protocol, and data analysis. Reverse transcription gets its own section. That placement is not accidental.

How qPCR Works

A qPCR reaction contains template DNA, a thermostable polymerase, dNTPs, magnesium, buffer, target-specific primers, and a fluorescent reporter. Thermal cycling proceeds as in conventional PCR, but fluorescence is read at every cycle.

The two detection chemistries

Intercalating dyes (SYBR Green I, EvaGreen) bind any double-stranded DNA and fluoresce when bound. They are inexpensive and require no probe design, which makes them good for assay development and screening. The cost is specificity: the dye cannot distinguish your amplicon from a primer dimer or a non-specific product. A melt curve at the end of the run is mandatory, not optional, and multiplexing is effectively impossible.

Hydrolysis probes add a third oligonucleotide, dual-labelled with a fluorophore and a quencher, that anneals between the primers. Polymerase 5′ exonuclease activity cleaves it during extension, separating reporter from quencher and releasing signal. Specificity is much higher because signal requires three independent hybridisation events. Multiplexing works, using spectrally distinct fluorophores. The trade-off is cost and design effort.

Reading the output

The amplification curve has three phases: a flat baseline where product is below detection, an exponential phase, and a plateau where reagents deplete. All quantification happens in the exponential phase. Endpoint fluorescence at the plateau carries almost no information about starting quantity, which is precisely why endpoint PCR is qualitative and qPCR is not.

Cq is where the curve crosses the threshold. Under ideal conditions with perfect doubling each cycle, a 10-fold difference in starting template shifts Cq by 3.32 cycles. A standard curve of serial dilutions converts Cq into absolute copy number; the ΔΔCq method compares Cq against reference genes for relative expression without a standard curve.

How RT-qPCR Works

Everything above still applies. It is preceded by cDNA synthesis.

Reverse transcriptase, an RNA-dependent DNA polymerase, synthesises a DNA strand complementary to the RNA template. The enzymes in common use are derivatives of Moloney Murine Leukemia Virus (MMLV) and Avian Myeloblastosis Virus (AMV), and increasingly engineered thermostable variants that stay active at 55–60 °C. That thermostability matters more than it sounds: many viral RNA genomes have extensive secondary structure that a mesophilic enzyme stalls on, and running the RT step hot melts it out.

Priming strategy is a real decision

How you prime the RT reaction changes your answer.

PrimerPrimes whatStrengthWeaknessOligo-dTPolyadenylated RNA, from the 3′ endSelective for mRNA; long cDNAFails on degraded RNA, non-polyadenylated targets, prokaryotic mRNA; under-represents 5′ ends of long transcriptsRandom hexamersAll RNA species, everywhereHigh total yield; tolerant of secondary structure and degradationrRNA is 80–85% of total RNA, so most cDNA is rRNA; dilutes low-abundance targetsGene-specificOne target onlyMaximum sensitivity for that target; used in one-stepcDNA cannot be reused for other genes; high GSP concentrations can inhibit the downstream PCROligo-dT + random blendBothBalanced coverage; common default for two-step eukaryotic workSlightly lower per-target yield than gene-specific

The magnitude of this effect is easy to underestimate. Ståhlberg and colleagues found random hexamers overestimating mRNA copy number by as much as 19-fold relative to other priming methods. Other work comparing primer and temperature combinations found 2- to 5-fold differences between the worst and best. None of that variation exists in a straight qPCR run.

The practical rule: pick one strategy and apply it identically to every sample in the experiment, calibrators included. Relative quantification tolerates a systematic bias. It does not tolerate an inconsistent one.

One-step versus two-step

One-step puts reverse transcription and amplification in a single closed tube with gene-specific priming. Fewer handling steps, dramatically lower contamination risk, better suited to automation and to diagnostic workflows. The cDNA is consumed in the reaction, so you cannot go back and test another target from the same synthesis, and you cannot easily troubleshoot which step failed.

Two-step syntheses a cDNA pool first, then aliquots it into separate qPCR reactions. You can assay many targets from one RT, archive the cDNA at −20 °C, and dilute out inhibitors carried through from extraction. The cost is more pipetting, more open tubes, and more opportunity for cross-contamination.

Diagnostics and high-throughput screening generally go one-step. Multi-gene expression studies generally go two-step. That maps directly onto product format — a single-reaction, all-in bead like the EVIK™ RT-qPCR Master Mix DxSpheres is built around the one-step model, where the entire reaction except primers, probes, and template is already in the tube.

Side-by-Side: qPCR vs RT-qPCR

qPCRRT-qPCRStarting templateDNARNAEnzymesDNA polymeraseReverse transcriptase + DNA polymeraseExtra stepNonecDNA synthesisTypical run time1–2 hours1.5–2.5 hours (one-step)Template stabilityHigh; DNA is robustLow; RNA degrades readily, RNase everywhereMain variability sourcePipetting, inhibitorsThe RT step, by a wide marginEssential extra controlNTCNTC and no-RT controlgDNA contaminationNot applicableSerious problem; needs active controlTypical targetsBacteria, DNA viruses, CNV, genotyping, GMO, ChIPGene expression, RNA viruses, miRNA, viral load

Choosing Between Them

The choice is dictated by the target's nucleic acid, not by preference. Some worked cases:

Bacterial detection — genomic DNA target, so qPCR. Unless you are asking whether the organism is metabolically active, in which case you want mRNA and therefore RT-qPCR. Presence and viability are different questions with different assays.

SARS-CoV-2, influenza A, HIV, hepatitis C, norovirus — all RNA viruses. RT-qPCR, no alternative.

Hepatitis B, HPV, herpesviruses, adenovirus — DNA viruses. Straight qPCR.

Gene expression — mRNA, so RT-qPCR. This is the largest single application of the technique.

Genotyping, SNP detection, copy number variation — genomic DNA. qPCR.

microRNA — RNA, but short enough that standard priming fails; requires stem-loop RT primers or polyadenylation-based methods. Still RT-qPCR, with modifications.

Retroviral integration — DNA, because integrated provirus is DNA. A common trap, since the virus itself is an RNA virus.

One useful check: if the assay must distinguish a live organism from residual nucleic acid after treatment, DNA-based qPCR will detect both. RNA-based detection is closer to a viability signal because mRNA turns over quickly, though it is not a clean proxy either.

Why RT-qPCR Is Harder to Get Right

Three problems belong to RT-qPCR alone.

The RT step is not quantitative by default

The convenient assumption is that RNA converts to cDNA proportionally, at a fixed ratio, uniformly across transcripts. It does not. Conversion efficiency varies with the enzyme, the primer, the input amount, and the target sequence, and the relationship between input RNA and cDNA yield is frequently non-linear across a dilution series.

The practical consequence: if your experiment requires varying RNA input across samples, verify that your RT protocol gives a linear response over that range before you trust any absolute number. For relative quantification against reference genes, consistency across samples matters more than absolute efficiency — which is why identical RT conditions for every sample is non-negotiable.

RNA is a difficult analyte

RNA is single-stranded, chemically less stable than DNA, and RNases are ubiquitous, robust, and do not require cofactors. Degraded RNA raises Cq and biases oligo-dT priming badly, because the poly-A tail may be separated from the region your amplicon targets.

Assess RNA integrity before committing to a run. Report the method and the metric. MIQE expects it, and a reviewer will ask.

Genomic DNA gives a false positive that looks real

Your primers usually amplify the DNA sequence as readily as the cDNA. Any gDNA surviving extraction is an amplifiable template, and the resulting signal is indistinguishable from genuine transcript.

Three defences, best used together:

1. DNase treatment of the RNA prep before reverse transcription.

2. Intron-spanning primers, so gDNA either fails to amplify or produces a distinguishably larger product. Not available for single-exon genes or processed pseudogenes.

3. A no-RT control (−RT) — the full reaction minus reverse transcriptase. Any amplification here is gDNA, full stop.

The −RT control is the one people skip and should not. It is the only direct measurement of how much of your signal is not RNA. In a one-step format it requires a separate reaction without the RT enzyme, which is inconvenient and still worth doing during assay validation.

Controls: What Each Assay Requires

Both:

• NTC (no-template control) — water instead of sample. Catches reagent contamination and primer dimers.

• Positive control — known template, confirms the reaction works when the sample is negative.

• Standard curve — for absolute quantification, and for measuring efficiency.

RT-qPCR additionally:

• No-RT control — as above.

• Internal amplification control — a spiked exogenous target that flags inhibition. Essential in diagnostics, where a false negative from an inhibited reaction is a clinical event.

• Reference genes — plural. Normalising to a single reference gene without validating its stability is a specifically named MIQE failure. Aim for two or three validated against your sample set.

Validation Criteria

The numbers a reviewer or auditor will look for, and which apply to both assays:

Amplification efficiency: 90–110%. Calculated from the slope of the standard curve as E = 10^(−1/slope) − 1. A slope of −3.32 is 100% efficiency. Below 90% suggests inhibition or poor primer design; above 110% usually means non-specific product or a pipetting error in the dilution series, not a polymerase exceeding physics.

R² > 0.98 across the standard curve, over at least five 10-fold dilutions.

LOD and LLOQ, determined empirically. The limit of detection is not simply the lowest standard that gave a signal; it is the concentration detected in ~95% of replicates. LLOQ is higher — the lowest concentration you can put a number on with acceptable precision.

Melt curve, for dye-based assays. A single sharp peak. Shoulders or low-temperature peaks mean non-specific product or dimers.

Technical replicate Cq spread within about 0.5 cycles. Wider means a pipetting problem.

For RT-qPCR, run efficiency validation on the full workflow from RNA, not on a cDNA dilution series. A cDNA dilution series measures the qPCR half only and will look excellent while the RT step quietly misbehaves.

Where Lyophilized Master Mixes Change the Picture

Both assays are conventionally run from frozen liquid reagents, which imposes real constraints: cold chain, freeze-thaw degradation, and a multi-component pipetting step that is the largest source of well-to-well variation in most labs.

Freeze-drying the reaction mix addresses all three. The relevant published work is encouraging. Fully premixed RT-qPCR reactions for SARS-CoV-2 detection have been lyophilized and stored at ambient and elevated temperatures without loss of activity, with disaccharide lyoprotectants stabilising the dried reactions against temperatures up to 50 °C for at least 30 days. Commercial lyo-ready RT-qPCR formulations report 24 months of ambient stability post-lyophilization with a ±0.5 Cq variance specification against a reference.

Two things make this harder than it looks:

Glycerol has to go. Enzymes are conventionally supplied in glycerol-containing storage buffer, and glycerol does not freeze-dry — it depresses the glass transition temperature and leaves a syrupy residue instead of a cake. Lyophilizable formulations require glycerol-free enzyme preparations.

Reverse transcriptase is the fragile component. It is generally less thermostable than Taq and takes more damage from the freezing and dehydration stresses of lyophilization. An RT-qPCR bead is a harder formulation problem than a qPCR bead, and the excipient system carries more of the load. This is exactly what excipient screening exists to solve — an EVIK™ Excipient Selection Plate runs a curated set of stabilisers, buffers, and surfactants in a 96-well format so compatibility is established empirically rather than assumed.

Where this matters in practice: point-of-care and field testing, where cold chain does not exist; multi-site studies, where lot-to-lot and site-to-site reagent variation confounds results; and any workflow where reducing pipetting steps meaningfully cuts error. EVIK™ Master Mix DxSpheres are single-reaction beads containing everything but primers, probes, and template — reconstituting in seconds and covering both qPCR and RT-qPCR from the same product line. If your workflow is isothermal rather than thermocycled, the same logic applies to RT-LAMP master mix beads.

The upstream step benefits too. RNA extraction is a major contributor to RT-qPCR variability, and stabilised, pre-measured DNA/RNA extraction MagSpheres remove one more manual dispensing step from a workflow that has too many.

Frequently Asked Questions

Is RT-qPCR the same as real-time PCR?

No, and this is the most common confusion in the qPCR vs RT-qPCR comparison. Real-time PCR is qPCR. RT-qPCR is reverse transcription qPCR — real-time PCR preceded by RNA-to-cDNA conversion. The RT stands for reverse transcription, never real-time.

Can I run RNA on a qPCR assay directly?

No. DNA polymerase cannot use RNA as a template. Without reverse transcription, you will get no amplification, or you will amplify contaminating genomic DNA and misread it as signal.

Can a qPCR master mix be used for RT-qPCR by adding reverse transcriptase?

Sometimes, but buffer conditions optimal for the polymerase are not necessarily optimal for the RT, and the two enzymes have different magnesium and salt preferences. Purpose-built one-step formulations are balanced for both. Improvising usually costs sensitivity on low-copy targets.

Which is more sensitive?

Neither inherently — the amplification chemistry is the same. In practice RT-qPCR often shows lower effective sensitivity for a given copy number, because incomplete cDNA conversion means not every RNA molecule becomes an amplifiable template.

Do I need a standard curve every run?

For absolute quantification, yes, or a validated calibrator traceable to one. For relative quantification by ΔΔCq, efficiency needs establishing during validation but not repeating every plate — provided reagents, instrument, and protocol are unchanged.

What Cq value counts as positive?

There is no universal cutoff. It depends on the assay's validated LOD, the sample matrix, and the total cycle number. A late Cq near the NTC boundary needs replicate confirmation and a check that the amplification curve has a genuine exponential shape rather than drift.

Why is my no-RT control amplifying?

Genomic DNA in the RNA prep. Add or extend DNase treatment, redesign primers to span an intron if the gene structure allows, and re-verify. Do not proceed on the assumption that the contribution is small enough to ignore.

The Practical Takeaway

Framed as qPCR vs RT-qPCR, the comparison sounds like a choice. It mostly is not — your target's nucleic acid decides for you. The choice you actually control is how carefully you handle the extra step, and whether you build the controls that make its variability visible.

If you are running qPCR, standard-curve efficiency and specificity are the things to guard. If you are running RT-qPCR, everything above still applies, plus RNA integrity, a consistent priming strategy across every sample, and a no-RT control you actually run rather than intend to.

Reagent format is a lever on all of it. Fewer pipetting steps means less variance, and ambient-stable reagents remove freeze-thaw as a variable entirely.

Evik Diagnostic Innovations manufactures preformulated, single-reaction lyophilized reagent beads for qPCR, RT-qPCR, and RT-LAMP workflows, backed by 15 years in diagnostic assay development and freeze-drying. Browse the full product range, read the technical FAQs, or contact the team about custom formulations for your assay.