Three biological replicates per condition, each assayed in technical triplicate on the qPCR plate. Include vehicle and treated arms plus a calibrator (control) condition for ΔΔCq. Load equal RNA mass (500 ng) into each RT reaction. Randomize sample-to-well placement and include inter-plate calibrators if more than one plate is used. Run target and reference assays on the same plate/run when possible to avoid run-to-run variation. Pre-validate primers with a 5-point 10-fold standard curve.
BSL-2 for human cell-derived material until RNA is purified; standard PPE (gloves, coat, eye protection). SYBR Green and intercalating dyes are potential mutagens — handle with gloves and dispose as chemical waste. Keep a nuclease-free RNA workspace separate from PCR-amplicon/post-PCR areas to prevent amplicon carryover contamination; never open post-PCR plates in the prep area. Dispose of sealed plates without reopening.
No-template control (NTC, water instead of cDNA) on every assay to detect contamination/primer-dimer. No-RT control per RNA sample to detect genomic-DNA amplification. Inter-run calibrator (a fixed reference cDNA) on every plate for multi-plate normalization. Positive control: a sample known to express the target. Reference-gene controls: ≥ 2 validated stable genes (geNorm M < 0.5, pairwise V2/3 < 0.15). Vehicle/untreated calibrator defines the 1× baseline for ΔΔCq.
Validated assays give standard-curve efficiency 90–110%, R² ≥ 0.99, single melt peaks, and technical-triplicate Cq SD ≤ 0.25. NTCs should be undetermined or ≥ 35 Cq; no-RT should be ≥ 5 Cq later than +RT. A true 2-fold induction appears as a ≈ 1 Cq decrease after reference normalization. Reference genes should vary < 0.5 Cq across conditions.
To measure relative expression of a target transcript in human cells across treatments by reverse-transcribing equal RNA mass to cDNA, amplifying with validated, single-amplicon SYBR Green primers, and normalizing to the geometric mean of two stably expressed reference genes selected by geNorm. The objective is MIQE-compliant quantification with documented amplification efficiency (90–110%) and clean melt curves.
Independent variable: treatment/condition. Dependent variable: normalized relative expression (fold-change vs calibrator). Controlled variables: RNA input mass (500 ng), cDNA dilution, primer concentration (0.3 µM), master-mix lot, cycling program, plate, and operator. Reference-gene set is held constant once validated. Amplification efficiency is a measured covariate that must fall in 90–110% for valid ΔΔCq.
Treatment will change target mRNA abundance by a biologically meaningful fold-change (e.g., ≥ 2-fold) relative to vehicle, detectable as a reproducible ΔΔCq shift when normalized to the geometric mean of two reference genes whose expression is stable (geNorm M < 0.5) across the conditions tested.
Average technical triplicates per assay. Compute ΔCq = Cq(target) − geometric-mean Cq(reference genes). Compute ΔΔCq = ΔCq(sample) − ΔCq(calibrator). Relative expression = 2^(−ΔΔCq), or use efficiency-corrected Pfaffl when efficiencies differ. Use qbase+, LinRegPCR, or instrument software; set a fixed threshold/baseline across the run. Report MIQE-required metadata (efficiency, RIN, primer sequences).
Late/no amplification: low cDNA or degraded RNA — confirm RIN and re-run RT; check primer efficiency. NTC amplifies: contamination or primer-dimer — use fresh aliquots, redesign primers, lower primer to 0.2 µM. Double melt peak: nonspecific product — raise anneal temp to 62 °C or redesign. High triplicate SD (> 0.5 Cq): pipetting/bubbles — use a master mix, centrifuge plate, calibrate pipettes. Reference genes unstable: re-run geNorm and choose a different reference set.
Perform statistics on log2 fold-change (ΔCq) values, which are approximately normally distributed, not on raw 2^(−ΔΔCq) ratios. Compare two groups by unpaired two-tailed t-test or ≥ 3 groups by one-way ANOVA with Tukey/Dunnett correction (α = 0.05); n = 3 biological replicates. Report mean fold-change with 95% CI. For multiple targets apply Benjamini–Hochberg FDR. With n = 3 and CV ~10% the design detects ~1.5-fold changes at 80% power; increase n for smaller effects.