Skip to content

Formulas

Formulas

Formulas are composed of macros on both sides, which are connected by '~'. It looks like the formula syntax from R: Y ~ X. This basically determines how the plots are generated.

However, we don't support operators like it from R. So we only have Y ~ X, but not Y ~ X1 + X2. In order to implement this, we have to define a macro to add X1 and X2 up.

We have three types of macros:

  1. continuous: extracts continues values from a variant, such as allele frequency, depth, etc.
  2. categorical: extracts categorical values from a variant, such as variant type, genotype, etc.
  3. aggregation: aggregates some values from a group of variants, such as mean of allele frequency from chromosome 1. You can also add a filter to aggregation by COUNT(1, filter=AAF[0.05, 0.95], group=CHROM)

See section Macros for details.

Tip

If you have both aggregations for Y and X, you can omit the group for either. The one without group will automatically use the group that is specified to the other one.

Predefined figure types for different combination of Y and X

Y X default figure type other available figure types
aggregation aggregation scatter -
aggregation categorical col pie
aggregation 1 pie col
anything other than aggregation aggregation not available -
categorical categorical bar pie
continuous categorical voilin boxplot/histogram/density/freqpoly
categorical 1 pie bar
categorical continuous (not 1) not available -
continueous continuous scatter -

geoms from ggplot2 used to plot for different figure types

figure type geom
scatter geom_point
col geom_col
pie geom_col with coord_polar
bar geom_bar
violin geom_violin
boxplot geom_boxplot
density geom_density
freqpoly geom_freqpoly