Adding Custom Trait Colors¶
Auspice uses a number of default color schemes to color the tree using meta data or values that the augur pipeline computed. In some cases these defaults are not suitable for particular type of data, and you’d like to use your own color schemes.
To specify a mapping between discrete trait values and colors, you can pass a tab-delimited file to
augur export v1/
augur export v2 using
Note that it’s not currently possible to specify color schemes for nucleotides, amino acids, or continuous data.
The file should contain 3 columns separated by tab characters. The first columns specifies the trait (e.g. country, host, …), the second columns specifies the value of the trait, and the third the associated color in hexadecimal notation.
country thailand #511EA8 country vietnam #4928B4 country singapore #4334BF country french polynesia #4041C7 country american samoa #3F50CC country fiji #3F5ED0
The hexadecimal notation starts with a #, followed by two letters coding for red, green, blue values. If you need advice or inspiration for colors, have a look at colorbrewer. Pay particular not to mix tabs and spaces for column separation. The line
country american samoa #3F50CC
for example has a tab between “country” and “american samoa” while “american” and “samoa” are separated by a space (because that happens to be the how this place is called). “samoa” and “#3F50CC” are again separted by a tab.
Augur matches geographic location in lower case, so the casing doesn’t matter. Beyond that, make sure your trait (e.g. ‘country’) and the values (e.g. ‘thailand’) exactly match what’s in your metadata!
Another reason you may want to use colors is to preserve ordering of your values. For example, if you have a trait
age_range with values
>65, you’d probably like them to appear in this order in the filter and legend.
By default, auspice would sort these alphabetically, so you’ll get them in a strange order. However, if you specify the order by assigning colors, the order will be preserved in auspice:
age_range <18 #511EA8 age_range 18-30 #4928B4 age_range 31-65 #4334BF age_range >65 #4041C7