Fig 1.
The wide versus tidy data format for time-dependent data.
In the wide (spreadsheet-like) data format, the first column holds the time data, and the other columns represent data from different objects (A, B, and C). In the tidy format, each column is a single variable, and each row is an observation. Note that the time data are repeated for each of the objects A, B, and C. The column id in the tidy data format identifies the condition. This column is automatically added and filled with a value of 1 when wide data are converted to tidy format. When multiple CSV files are uploaded with wide data, this column is filled with the filename.
Fig 2.
Different plots generated in PlotTwist with the same data.
An ordinary line plot, showing the individual data as lines (A) and a heatmap-style presentation (B) of the (same) example data. Both graphs are annotated with bars on top of the plot to display an activating and an inhibiting perturbation.
Fig 3.
Composition and application of the color-blind-friendly palettes that are available in PlotTwist.
(A) An overview of the color composition of the qualitative color palettes that were designed to have colors that are distinct to all people, including those with a color vision deficiency. The RGB code for each of the colors is indicated in hexadecimal code, preceded with a hashtag. (B) A realistic application of the color palettes shown in the upper panel for the unique labeling of 6 different lines. RGB, Red Green Blue.
Fig 4.
Different annotation styles that are available in PlotTwist.
From left (A) to right (C), increasingly explicit labeling is used to explain the data. The data are the same for each plot and show the relative activity of the proteins Cdc42, Rac1, and RhoA in response to treatment with sphingosine 1-phosphate [14]. (A) A standard legend is used to explain the color used for the lines. (B) Direct annotation of the lines is used by placing a label next to the data. In addition, the color of the label is the same as the color of the data. (C) A fully annotated plot with clear labeling of the axes, direct annotation of the data, and a visualization of the treatments at the different times.
Fig 5.
Result of the clustering of data set Example 2 in PlotTwist.
The data set was subjected to clustering with Euclidean distance and Ward.D2 linkage, using an optimal number of 4 clusters as inferred from the Calinski Harabasz index. The plots show the data allocated to a cluster and how the data of the 3 Rho GTPases (RhoA, Rac, and Cdc42) are distributed over the 4 different clusters.