mvSuSiE PIP and effect plots — mvsusie

Creates PIP and effect-size plots showing effect estimates and significance across traits. A z-score plot is also created when z-scores are available.

Usage

mvsusie_plot(
  fit,
  chr = 1,
  pos = seq(1, length(fit$variable_names)),
  markers = fit$variable_names,
  outcomes = fit$outcome_names,
  poslim = range(pos),
  lfsr_cutoff = 0.01,
  sentinel_only = TRUE,
  cs_plot = names(fit$sets$cs),
  add_cs = FALSE,
  conditional_effect = TRUE,
  sort_by_cs = FALSE,
  cs_colors = c("#1f78b4", "#33a02c", "#e31a1c", "#ff7f00", "#6a3d9a", "#b15928",
    "#a6cee3", "#b2df8a", "#fb9a99", "#fdbf6f", "#cab2d6", "#e6ab02", "gray", "#66c2a5")
)

Arguments

fit: The mvSuSiE fitted model.
chr: The chromosome number.
pos: The positions of the genetic markers. It should have the same length as fit$variable_names.
markers: The names of the genetic markers (usually SNPs).
outcomes: The names of the outcomes.
poslim: The range of positions to show in the PIP plot.
lfsr_cutoff: The significance level for lfsr. The default is 0.01.
sentinel_only: If TRUE, only plot the sentinel marker for each CS. If FALSE, plot all markers in each CS.
cs_plot: The CSs included in the plot. The default is to show all CSs.
add_cs: If TRUE, add colored dots to the top of the effect plot showing CS membership.
conditional_effect: If TRUE, plot the conditional effect. If FALSE, plot the marginal effect. conditional_effect = TRUE is recommended.
sort_by_cs: If TRUE, group markers by CS on the x-axis of the effect and z-score plots. This prevents CSs from interlocking when their members are interleaved in genomic position. The default is FALSE (genomic order).
cs_colors: The color palette for CSs.

Value

The output includes the PIP plot, effect plot, z-scores plot (if z-scores are available in fit), and the table of effect estimates at sentinel markers.

Examples

# See the "mvsusie_intro" vignette for examples.