Relative quantification estimates the original quantity of one or more targets in unknown samples, relative to reference gene targets. This analysis makes use of Cq Calling analysis from amplification curves, and can be used to determine changes in gene expression levels of your gene of interest.
Performing Relative Quantification
To demonstrate relative quantification, we will use an example data file. Open the “MyGo Pro Relative Quantification.ppf” file in the walkthrough data folder. Now click the Profile tab, where we can see that a simple cycling profile has been run. All quantification analyses require a single cycling phase to be present in the profile. This can be a normal cycling phase or a touchdown phase.
Click on the Samples tab to display the samples and targets layout of the project:
This is the same layout prepared in the section on Samples Setup. Scroll through the Wells as Table display to familiarise yourself with the layout - we have two samples (“Sample A”, and “Sample B”), two genes of interest (GOI) (“Gene 1” and “Gene 2”) and one reference gene (“Reference 1”). For each sample we have two replicates of each gene. In this example we do not have any standards - as we will see later this will lead us to perform a simple delta-delta-Cq analysis. It is also possible to define standards curves to allow for efficiency correction.
Absolute Quantification Analysis
Now click on the Analysis tab. The Absolute Quantification analysis stage is selected at the top of the window:
This display should be familiar from the section on Absolute Quantification. The Results as Table display shows the sample and target layout. Since we have one dye, FAM, we have one dye column, showing the target name and type, Cq value, positive/negative call, and estimated quantity associated with that dye in each well. As can be seen, each dye and target is analysed separately.
When performing relative quantification, the Absolute Quantification display is used to check Cq calls, display amplification curves, edit Cq calling settings and view notes, exactly as when performing any other quantification. However the quantity columns of the results are not as useful - they simply give the quantities as they would be estimated in an absolute quantification. To see the relative quantification results, we will proceed to the next stage of the analysis.
Relative Quantification Analysis
Click on the Relative Quantification tab to display relative quantification settings and results. First, click the Samples tab in the lower pane, to display the samples table:
This table displays the samples defined in the experiment.
This is also where we define which sample to use as a Calibrator. This is a reference sample used for correcting ΔCq values to produce a ΔΔCq value, and hence a normalised ratio - this is described in more detail later. To select a Calibrator, click on the checkbox in the Calibrator column of the table. Only a single Calibrator can be defined in each experiment. In this experiment Sample B has been chosen as the calibrator. It’s Ratio results have now been set to 1.
Samples and Targets Tabs
These tabs allow you to see which samples and targets have been defined in the experiment. Names can be edited here if the user wishes.
Group by Samples
Grouping by samples will show you the relative quantification between Gene 1 and 2 from Sample A and Sample B as shown below:
Group by Targets
Grouping by targets will show you the relative quantification of Sample A and Sample B for Gene 1 and 2 as shown below:
Graph Selected Samples Only
The check box at the bottom of the screen allows the user to only display the sample or target which has been selected in the table. For example, when in the Samples tab and Sorting by targets the graph will show relative expression for Gene 1 and 2 for Sample A only as shown below:
Similarly, selecting one or more samples or targets and clicking Graph selected samples only will filter the chart to display only RQ values associated with the selected targets.
Click on the Analysis Results tab in the lower area:
This display shows the results of taking the Cq values and quantifiers as shown in the previous Cq and Absolute Quantification tab, and applying the method of relative quantification.
The results table has a row for each combination of a sample and a target. So for example, the first row in our example project relates to the quantity of “Reference 1” in “Sample A”, the next relates to quantity of “Gene 1” in “Sample A”, and so on.
The first column shows the name and colour of the relevant sample, and the second column shows the name and colour of the relevant target.
The third column shows the mean and standard deviation of all efficiency-corrected Cq values for amplification of the relevant target in the relevant sample. So in the example experiment, the mean Cq of “Reference 1” in replicates of “Sample A” is 21.02, and the standard deviation is 0.09 (you may need to resize the window or the table column to see the standard deviation, which is displayed after the +/- symbol).
The Cq values displayed in the table are all corrected for the specific efficiency of the target they relate to. The correction ensures that the Cq values displayed are consistent with an ideal efficiency of 2 and so are comparable to each other. This correction is based on the quantifier for each target - so if a given target has a standards curve, the efficiency derived from that standards curve will be used to correct all Cq values for that target so that they are consistent with an ideal efficiency of 2. If no standards curve is present for a target, the efficiency value entered in the Quantifier display will be used instead - this is described in more detail in the section on Absolute Quantification.
Since our example data has no standards defined, all Cq values are corrected based on the default efficiency of 2. This value can be edited for each target in the same way as for an absolute quantification. However for relative quantification analysis, only the efficiency value is used.
Returning to the Analysis Results tab as shown above, we can see that the next column of the table displays ΔCq values. These indicate the Cq of each GOI relative to the reference target. So in the example data above, we can see that the GOI “Gene 1” in “Sample A” has a mean Cq of 26.34. Reference 1 has a mean Cq of 21.02 in “Sample A”.
ΔCq = Cq of target of interest - Cq of reference target
ΔCq = 26.34 - 21.02 = 5.32
This indicates that “Gene 1” is present in “Sample A” at a level that causes it to amplify 5.32 cycles later than Reference 1, when corrected for ideal amplification. Ideal amplification implies a doubling per cycle, and so this represents a quantity of “Gene 1” that is 2 to the power of -5.32 = 0.02501 times that of Reference 1*,* as displayed in the Ratio column. The Ratio Min. and Ratio Max. columns display the same calculation performed using a range of ΔCq given by +/- 3 standard deviations.
Note that in this experiment, there is only a single reference target- however as described in Samples Setup, multiple targets may be marked as reference targets. In this case, the ΔCq values of GOI are calculated relative to the mean Cq value across all GOI. This can give a more stable reference than use of a single reference target.
The next column of the table displays ΔΔCq values. Each ΔΔCq value indicates the ΔCq value of a target of interest relative to the ΔCq value of the same target of interest in the reference sample. In the example data above, we use “Sample B” as the reference sample, which is referred to as the calibrator. Hence the ΔΔCq values for GOI in “Sample A ” are -1.19 for Gene 1 and 0.02 for Gene 2. You may choose a sample to use as the calibrator by clicking on the checkbox in the Calibrator column of the Samples table, as described earlier.
The final three columns show the normalised ratio of the targets of interest (“Gene 1” and “Gene 2”), when corrected by both the reference targets, and the calibrator.
The Analysis Results table shows all values with an error estimate. For E Corr. Cq, ΔCq and ΔΔCq this is in the form of:
mean ± standard deviation
For the Ratio and Norm. Ratio (normalised ratio), the values given are mean, and then a minimum and maximum value based on 3 standard deviations. The mean/min/max format is used because the error range for RQ is not symmetrical (due to conversion from log-based Cq value to linear quantity value).
The contents of the Analysis Results table can be exported to a Comma Separated Value (.csv) or a PDF (.pdf) file using the Export… button.
Relative Quantity Chart
The Relative Quantity Chart shows the same RQ data as the Analysis Results table, but in a bar chart format.
Each bar represents the mean RQ of a GOI for a particular sample. Error bars show the minimum and maximum RQ value, again calculated using 3 standard deviations.
By default, the bars are grouped by sample, and coloured by target. The samples and targets are ordered the same way as in the tables.The chart has standard controls for zooming and exporting data.In addition, the chart has a control at the right hand side, showing a bar and a mouse cursor. When this control is enabled, bars can be selected by clicking on the chart. The selection is shared with the Analysis Results table.
Relative Quantification Quantifiers
Relative quantification uses the same quantifiers as absolute quantification, however only the efficiency values are used, not the absolute quantities. The efficiency value can be specified manually, based on an imported standards curve, or based on standards in the experiment itself, as usual.
Click the Absolute Quantification tab at the top, then click the Quantifiers tab in the top left pane to display the quantifiers.
In the screenshot above, we can see that “Reference 1” is selected, and an Efficiency: value of 2 is entered in the number control.
Click on the Selected Target: drop down control, and select “Gene 1”. We can see that this is also assigned an efficiency of 2:
We can see that each target can have its own efficiency value assigned - using the correct efficiency for a target is important for accurate quantification. Accuracy can be further improved by importing a standards curve from another experiment, or ideally running standards for each target in each experiment.
This concludes the section on Relative Quantification**_._** More details on Cq calling and quantifiers may be found in the section on Absolute Quantification.