Input file format requirements:
- The input file must be in tab-delimited text or excel format.
- List temperatures in the first column (do not leave blank columns).
- List sample names in the first row.
- Sample names must be text or text and numbers, not just numbers.
- All control samples should contain the word "control" in the sample name.
Select a data file for analysis and click "Generate Figures".
The resulting graphs show a plot of the raw data and filtered data in which poorly amplified samples are removed. The "Data quality cutoff" slider can be used to adjust the threshold used for this filtering step.
Select regions of the melt curves that will be used to normalize between individual samples. Ranges can be selected either by dragging the vertical lines on the plot or entering temperature values directly.
The two regions should be selected such that they include "background" but not "signal" as illustrated below.
The "Overlay" option can be used to shift all curves along the temperature axis such that they overlap at a point defined using the "Overlay position" slider. This can be used to consider only differences in curve shape rather than melting temperature.
Click "Generate Figures" to view the normalized data.
Enter a p-value cut-off used to identify samples likely to be different from controls. Significant curves will be displayed in blue.
In addition, clustering can be performed to divide the data into a user-defined number of groups based on curve similarity. This can be useful to identify samples with similar sequences. Click "Generate Figures" to view results and download output files.
|Variables.txt||Lists the user-defined variables used for analysis|
|Data_filtered||Data used to generate the plot on page 1|
|Data_normalized.txt||Data used to generate the plot on page 2|
|Data_dif.txt||Data used to generate significance plot|
|Data_clustering.txt||Assignment of samples to clusters|
|Data_significance.txt||Assignment of p-values to samples|