LinRegPCR

LinRegPCR is a program for the analysis of quantitative RT-PCR (qPCR) data resulting from monitoring the PCR reaction with SYBR green or similar fluorescent dyes. The program determines a baseline fluorescence and does a baseline subtraction. Then a Window-of-Linearity is set and PCR efficiencies per sample are calculated. With the mean PCR efficiency per amplicon, the Cq value per sample and the fluorescence threshold set to determnine the Cq, the starting concentration per sample, expressed in arbitrary fluorescence units, is calculated. See: Ramakers et al., NeuroSci Lett 2003; Ruijter et al., Nucleic Acids Research 2009.

FAQs

I tried to import a file into LinRegPCR : the error message was "invalid variant type conversion".
This error message indicates that the program is trying to convert a text that is not a number into a number. This most often happens because: 1: you gave the wrong input range or 2: the decimal separator in Excel is different from the one in Windows. LinRegPCR works with the decimal separator in Windows, so you have to set Excel to use the system separator (in Excel: options menu - international Tab in the options dialog).
After importing data I get the error message "floating point division by zero".
This error occurs when there are not enough data per sample to fit a straight line. This may be because no positive values are left after baseline subtraction by your PCR apparatus. Check your input file and remove the rows or columns with these 'empty' samples. Better still: leave the baseline correction to LinRegPCR (see: Ruijter et al. NAR 2009)
I did run LinRegPCR but in the output all the starting concentrations are 0.
This is because Excel only displays three decimal places. When you increase the number of decimal places or switch to scientific format you will see the starting concentrations are above 0. Except when no baseline can be determined or no amplification is present, then the program reports -999.
Why does LinRegPCR use the word 'baseline' in stead of 'background'?
We use the word baseline to describe the fluorescence that is observed before the amplicon specific fluorescence can be detected. Most PCR systems already use the word background for the fluorescence of a reference fluorochrome that is used to correct for experimental variations in sample volumes and well characteristics. In these systems baseline is used as we use it.
What is the unit of the N0 value?
The starting concentrations (N0) per sample are calculated in the unit of the Y-axis of the PCR amplification plot which are arbitrary fluorescence units. To convert this unit to a RNA concentration you need a calibration line of known concentration of the amplicon you are producing in the PCR.
LinRegPCR reports PCR efficiency values that range between 1 and 2. How must I interpret these values?
You are probably used to describe efficiency as a value between 0 and 1. To get these values you just subtract 1 from the efficiency that LinRegPCR reports. An efficiency of 1.85 reported by LinRegPCR can be read as an efficiency of 0.85 or 85%. We use PCR efficiencies between 1 and 2 because it makes the equations a lot easier to handle.
How is the starting concentration in LinRegPCR calculated?
LinRegPCR calculates a starting concentration (N0) per sample with the formula: N0 = Nq / (Emean^Cq) with the ^ symbol meaning 'to the power'. In this formula Nq stands for the fluorescence threshold set to determine Cq, which is the number of cycles needed to reach Nq. Emean is the mean PCR efficiency for the amplicon that is amplified in the current sample. The mean efficiency is used because the efficiency per sample is too variable to give reliable results. (see Karlen et al., 2007; Cikos et al., 2007)
I have always used the comparative Ct method to calculate the expression of a target relative to a reference gene. How do I do that with the results of LinRegPCR?
LinRegPCR gives you the starting concentrations (N0) of the target and the reference genes. When you have replicates per biological sample you first take the average of the N0 values in the target wells and in the reference wells. Then the relative expression is the ratio of these two averages. When you have only one measurement per amplicon, you calculate the ratio directly from the N0 values. You do not need to use the efficiency values and the Cq values to do this. These are only displayed in the output to give you a chance to do a comparative Ct calculation. The result will be the same. See the Equations in Box 1 in Ruijter et al., NAR 2009.
LinRegPCR gives a warning about 'noisy samples'. How do I recognise a noisy sample?
A noisy sample is defined as a sample in which the data points do not show a continuous increase in the Window-of-Linearity. You recognise a noisy sample because in the W-o-L they have a point c that is above -or at the same level as- point c+1. Noisy samples are excluded from the calculation of the mean efficiency.
LinRegPCR reports a lot of baseline errors but the amplification curves show a straight log-linear phase.
LinRegPCR reports a baseline error when it is not possible to find a baseline value that leads to a straight continuous set of data points in the log-linear phase. However, the program also reports a baseline error when the remaining log-linear phase is to short. This may be because measurement noise causes the lower data points to be discontinuous: the fluorescence in cycle c is larger than that in cycle c+1. The baseline estimation only uses data in which Fc < Fc+1. In version 11.3 you can 'relax' this continuity criterion and allow jumps, as long as all data in the log-linear phase are around a straight line. This leads to less baseline-error samples but also leads to more variation between individual efficiency values.
I am wondering whether you can apply your window of linearity methodology to qPCR data obtained using a Taqman probe assay rather than the SYBR green assay? I can't think of any, as both result in fluorescence values but perhaps I have missed something?
The kinetics of the fluorescence of the Taqman probe is different from SYBR green. SYBR green is binding to dsDNA and is freed again at heating. So the fluorescence you see is proportional to the amount of DNA present at the end of each cycle. The Taqman probe binds to ssDNA, is digested by the polymerase and then becomes fluorescent. And stays fluorescent. So the fluorescence you see is an accumulation of the probe that has ever bound to the ssDNA. Therefore, this fluorescence increases more rapidly than the SYBR green fluorescence. However, on a log-fluorescence scale this amplifiaction curve becomes parallel to the curve that would have been observed when the reporter fluorescence was not cumulative. Therefore, the derived PCR efficiency is correct. Because the values in the Taqman curve are higher, its Cq value is too low. However, this bias is only dependent on the PCR efficiency and can thus be corrected. (see Tuomi et al. , Methods 2010)
Q.
A.

References

Tuomi et al. Bias in the Cq value observed with hydrolysis probe based quantitative PCR can be corrected with the estimated PCR efficiency value. Methods 50: 313-322, 2010
Ramakers et al. Assumption-free analysis of quantitative real-time PCR data. Neurosci Letters 339: 62-66, 2003
Ruijter et al. Amplification efficiency: linking baseline and bias in the analysis of quantitative PCR data. Nucleic Acids Research 37: e45, 2009
Karlen et al. Statistical significance of quantitative PCR. BMC Bioinformatics 8: 131, 2007
Cikos et al. Relative quantification of mRNA: comparison of methods currently used for real-time PCR data analysis. BMC Mol Biol 8: 113, 2007

Version History

LinRegPCR version 12.4 released July 2010

Bug fix in baseline estmation. Bug caused a crash but had no effect on estimated baselines.
Bug fix in import of Step-One Plus output: Rn column is used, delta-Rn column is ignored.

LinRegPCR version 12.3 released May 2010

Restricted application of Cq-shift correction to hydrolysis probe monitered qPCR with input of ds-DNA.

LinRegPCR version 12.2 released March 2010

Updated manual released.
Tuomi et al reference completed.

LinRegPCR version 12.1 released February 2010

Correction of an error in the display of flagged samples.
Exclusion of deviating PCR efficiency values was adapted to handle skewed distiributions.

LinRegPCR version 12.0 released February 2010

Update because of acceptance of the paper in Methods (Tuomi et al. )
This paper describes how to deal with cumulative fluorescence data like those obtained with TaqMan hydrolysis probes.

LinRegPCR version 11.5

Updated Rotor-Gene input format to include raw and baseline-corrected data
Raw data are exported with the "Excel data sheet" option of Rotor-Gene
The LinReg export format exports baseline-corrected data (constant baseline per sample)

LinRegPCR version 11.4

not released

LinRegPCR version 11.3

Updated the code to read the Step-One Plus format to allow for different output versions
Enabled a 'relaxed' baseline estimation for datasets with a short continuous log-linear phase due to measurement noise
Extended 'Legends' column in output with list of choices made by the User

LinRegPCR version 11.2

Added import formats to Read-from-Excel dialog:
- Step-One Plus system (ABI)
- LightCycler 480, 2 columns per sample
- Rotor-Gene (Corbett Reseach; LinRegPCR export)
Corrected program flow after detection of noisy samples
Enabled easier handling of baseline error samples

LinRegPCR version 11.1

Removed 'overflow' error that occurred sometimes when empty wells were included in the data

LinRegPCR version 11.0 released January 2009.

Version number was increased because of the acceptance of the paper in Nucleic Acids Reasearch.

Added input format for Stratagene systems (Format 1, Vertically Grouped data)
Added export of input data in ‘1 row per sample, 1 column per cycle’ format

LinRegPCR version 10.3

Detection of noisy datasets
Extended quality check with ‘noisy sample’
Removed error that lead to crash in the setting of the window-of-linearity in noisy datasets
Help on the handling of noisy data

LinRegPCR version 10.2

Corrected error in the setting of the Ct value
Corrected assignment of upper W-o-L limit for first amplicon group
Implemented full reset of the values when ‘determine baseline’ is pressed twice

LinRegPCR version 10.1

‘division by zero’ error that occurred when not enough data were present at low W-o-L settings was captured and handled
Freezing of the tab pages when a grouping error occurred was removed

LinRegPCR version 10.0 fully updated version. Released March 2008.

In this version the following features have been implemented:

import of raw data files
estimation of baseline fluorescence
new approaches to determine the window-of-linearity and fluorescence threshold
definition of sample groups per amplicon
calculation of the mean PCR efficiency per amplicon group
calculation of starting concentrations based on mean efficiency and individual Ct values
reports on data quality for each sample

LinRegPCR (automated content)

Download: LinRegPCR: analysis of qPCR data

If you wish to receive mails regarding important updates/bugs, please enter your email address below and press submit. The address will not be used for any other purpose