DATASETS

---

DATASET OVERVIEW

A dataset typically consists of three sections

• Annonations
• modifiers
• data

Let's see some example first:

## -- here is an example dataset of a barplot;
## lines start with # are annotations and will not be parsed by EvolView --
## -- the following are the modifiers --
## -- please use TAB to separate the modifiers and their values
## -- !legendTitle   or !title
!legendTitle	Example of barplots
!legendText	group 1,group 2,group 3
!legendStyle	circle
!colors	#028482,#7ABA7A,#B76EB8
!plotwidth	200
## -- annotation can be also put here --
## -- here comes the data --
## -- please use TAB to separate the columns --
A	8,13,5
B	10,20,7
C	8,9,7
D	20,5,20

And here is the resulting visualisation:

Annotation

Lines with the first character as '#' are annotations. Annotations are usually at the beginning of a dataset; however if desired, annotation lines can be placed anywhere in a dataset.

Modifiers

KA - pairs usually start with '!', and are separated by a single 'tab' character; KA pairs are used to specify or change certain default settings / behavior of a dataset, for example:

• to set the title or legend
• to change the width, height or opacity of the dataset
• in some cases, the attribute part of a KA-pair is optional.

Here are a list of KA-pairs common to all datasets; dataset-specific KA-pairs will be introduced later.

Key (case insensitive)	Value	Description
!Groups or !LegendText	comma separated text	Legend texts; for example 'group_a,group_b,group_c'
!LegendStyle or !Style	rect or circle or star	shapes to be plotted before the legend texts; default = rect
!LegendColors or !Colors	comma separated color codes or names	colors to be applied to the shapes specified by LegendStyle; for example 'red,green,yellow' ; note the number of colors should match the number of legend fields
!Title or !Legend	text	title of the legend; default = name of the dataset
!ShowLegends	0 or 1	0 : hide legends; 1 : show legends
!opacity	float number between 0 to 1	opacity of the dataset
!PlotWidth	integer > 0	pixel width of the dataset on canvas

Data

Data are usually tab-delimited three-column texts; the first two columns are often mandatory, the third one is often optional.

first column: the location

The first column dictates where the data to be plotted. It usually contains the name of a leaf node, or two leaf names separated by a ','.

• one single leaf name dictates that the data will be plotted on / next to / under the leaf or the branch connecting directly to the leaf node
• two leaf names, on the hand, dictates that the data will be displayed on the branch representing the last common ancestor of the two leaf nodes For example:
• chicken
• mouse,human

second column: data to be displayed

in bar plots, the data are the numbers representing the widths of the bars. see more examples in each of the datasets

third column: optional commands to change the default behavior of the input

see the examples in each of the datasets.