Skip to content
OpenReadout

Check a qPCR run

Use this after a qPCR run to see the Cq of every well, the efficiency and R² of a standard curve, and relative expression by ΔΔCq. It reads RDML files, Applied Biosystems .eds, Rotor-Gene .rex and LightCycler 480 .ixo.

Run it

$ openreadout analyze qpcr rdml-stepone-std.rdml --standard-curve
rdml-stepone-std.rdml (rdml, rdml): Standard Curve Example
instrument: Applied Biosystems StepOne™ Instrument
24 well x target records
well   sample                 target            task        Cq        Tm
A1     NTC_RNase P            RNase P           ntc        undet.(40)         -
...
A4     pop1_RNase P           RNase P           unknown     28.963         -
A5     pop1_RNase P           RNase P           unknown     28.839         -
...
B2     STD_RNase P_10000.0    RNase P           standard    26.874         -
...
C8     STD_RNase P_625.0      RNase P           standard    31.035         -


Cq: 21 determined, 3 undetermined, 0 no result (0 excluded)
...


standard curve RNase P: slope -3.4770, intercept 40.768, R² 0.9995, efficiency 93.9 % (15 wells, 5 levels)
note: target RNase P: amplificationEfficiency 93.91181 read as a percentage (1.9391 fold)
note: 3 reactions store a cq at or beyond the run's cycle count, which is how the exporting software writes "no Cq": reported as undetermined, the stored number kept as cq_stored
...

rdml-stepone-std.rdml is an RDML file written by StepOne Software, from the RDML R package’s examples (MIT). It is committed at fuzz/corpus/core_zip/rdml-stepone-std.rdml: one target (RNase P), five standards from 625 to 10000 copies in triplicate, two unknown populations and three NTCs. The output on this page is real, with long tables trimmed.

What it tells you

  • The table has one row per well and target. Cq is the value the file stores, as the instrument software called it. undet. marks wells the file says did not amplify; they are left out of averages.
  • Look at the NTC wells. The file stores 40 for them, the run’s cycle count, which is how StepOne writes “no Cq”. OpenReadout reports a Cq at or past the cycle count as undetermined, shows the stored number in brackets and keeps it as cq_stored in the JSON. Their curves rise about 3 % over baseline, against 257–331 % for the amplified wells.
  • The standard curve is a least-squares line of Cq against log10(quantity) over the standard wells. A slope of −3.32 means 100 % efficiency; efficiency = (10^(−1/slope) − 1) × 100, here 93.9 %.
  • When the file stores the vendor’s own fit, its slope, efficiency and R² are given next to ours (vendor_slope, vendor_efficiency_percent and vendor_r2 in the JSON). This file stores only the target’s efficiency, 93.91181, which matches.

Variations

Relative expression (ΔΔCq)

$ openreadout analyze qpcr eds-7500-abhd17c-ddct.eds --ddcq
...
ΔΔCq (reference: 18s; control: Lenvatinib/Vector)
sample                 target            n   meanCq    ΔCq     ΔΔCq      RQ
ABHD17C OE             ABHD17C            2   22.538  14.092   -5.614  48.988
Lenvatinib/ABHD17C OE  ABHD17C            3   22.995  14.398   -5.308  39.623
Lenvatinib/Vector      ABHD17C            3   27.145  19.706    0.000   1.000
Vector                 ABHD17C            3   27.192  19.530   -0.177   1.130

eds-7500-abhd17c-ddct.eds is a 7500 run with two targets, ABHD17C and the 18S reference, from Figshare (CC-BY-4.0). The file records 18s as the endogenous control and Lenvatinib/Vector as the calibrator, so no flags are needed; --reference TARGET and --control SAMPLE override them. Each sample and target gets ΔCq, ΔΔCq and RQ = 2^−ΔΔCq with its range. ABHD17C OE reads about 49 times the calibrator’s level. A file without a reference target gives a usage error that lists its targets:

$ openreadout analyze qpcr rdml-stepone-std.rdml --ddcq
error: usage error: ΔΔCq needs a reference target (endogenous control): pass --reference TARGET; this file's targets: RNase P
hint: Check the arguments with `openreadout help <command>`; `openreadout info FILE --json` shows what the file holds.

Recompute Cq from the curves

--cq computes a threshold Cq for every curve and compares it with the stored one:

$ openreadout analyze qpcr rdml-stepone-std.rdml --cq
...
our Cq vs vendor: 24 curves, 24 both with Cq, 0 both undetermined, 0 only vendor, 0 only ours; mean diff -7.369, median |diff| 6.909, max |diff| 28.629, within 0.5 cycles 0.042, r = 0.50760

The two agree closely only when the file records the threshold and baseline the software used. This StepOne RDML file records neither, so OpenReadout falls back to its own automatic threshold, and the difference above is a difference of method, not a reading error. Use the stored Cq for such files, and set --threshold and --baseline to match your software’s settings when you need the comparison. See qPCR formats.

Many runs

$ openreadout batch qpcr qruns/ --set standard_curve=true
2 data sets (2 ok, 0 failed)
path                         format  target   points  levels   slope  intercept      r2  efficiency_percent
...
qruns/rdml-stepone-std.rdml  rdml    RNase P      15       5  -3.477    40.7681  0.9995             93.9102
qruns/small.rdml             rdml    -             -       -       -          -       -                   -

small.rdml has a single standard well, so it gets no curve. --set ddcq=true gives one row per sample and target instead.

From an assistant

The MCP tool is openreadout_analyze with kind: "qpcr" and options such as standard_curve, ddcq, reference_targets and control_sample.

More