visualise graph of omop hierarchy called by omop_graph()

Usage

omop_graph_vis(
  graphlist,
  ggrlayout = "graphopt",
  palettebrewer = "Dark2",
  palettedirection = 1,
  edgecolour = "grey71",
  nodecolourvar = "domain_id",
  textcolourvar = NULL,
  nodealpha = 0.8,
  edgealpha = 0.3,
  edgewidth = 0.1,
  nodesizevar = "connections",
  nodesize = c(0, 6),
  nodetxtvar = "name",
  nodetxtangle = 0,
  nodetxtsize = 9,
  nodetxtnudgey = 0.3,
  nodetxtnudgex = 0,
  legendtxtsize = 18,
  titletxtsize = 18,
  titlejust = "centre",
  legendshow = TRUE,
  legendpos = "bottom",
  legenddir = "horizontal",
  legendcm = 3,
  plot = TRUE,
  saveplot = TRUE,
  filetype = "pdf",
  filenameroot = "omop_graph",
  filenamecustom = NULL,
  filepath = file.path("..//omopcept-plots"),
  canvas = NULL,
  width = 50,
  height = 30,
  units = "cm",
  titlecolour = "darkred",
  backcolour = "white",
  graphtitle = "omopcept graph",
  graphsubtitle = NULL,
  caption = NULL,
  captiontxtsize = 18,
  captionjust = "left",
  captioncolour = "black",
  messages = TRUE
)

Arguments

graphlist: list of edges and nodes created from omop_graph_calc()
ggrlayout: ggraph layout, default = "graphopt", also "tree" works well, more directional
palettebrewer: colour brewer palette, default='Dark2', other options e.g. 'Set1' see RColorBrewer::brewer.pal.info
palettedirection: palette direction, default=1, -1 for reversed
edgecolour: colour for lines joining nodes
nodecolourvar: column to specify node colour, default="domain_id" other options "vocabulary_id" "concept_class_id" "standard_concept"
textcolourvar: column to specify node text colour, default=NULL then set same as node_colour above. Other options "vocabulary_id" "concept_class_id" "standard_concept"
nodealpha: node transparency, default 0.8
edgealpha: edge transparency, default 0.3, #ggraph uses underscore edge_alpha but would mess up my consistency
edgewidth: edge width, default 0.1, #ggraph uses underscore edge_width but would mess up my consistency
nodesizevar: column to set node size, default="connections", uses num connections to a node
nodesize: modify node size range, default c(0,6), will modify size whether nodesizevar used or not
nodetxtvar: column to set node label, default="name"
nodetxtangle: node text angle, default=0, 90 gives vertical text
nodetxtsize: node text size, default=9
nodetxtnudgey: nudge_y text relative to points, default 0.3
nodetxtnudgex: nudge_x text relative to points, default 0
legendtxtsize: text size for legend, default=20
titletxtsize: text size for title, default=20
titlejust: title justification, "left","right", default "centre"
legendshow: whether to show legend, default TRUE
legendpos: legend position, default 'bottom'
legenddir: legen direction default = 'horizontal'
legendcm: legend size cm, default=3
plot: whether to display plot, default TRUE, note that large plots will not display well in R graphics window but do output well to pdf
saveplot: whether to save plot, default TRUE, note that large plots will not display well in R graphics window but do output well to pdf
filetype: output image file, default='pdf'
filenameroot: optional root for an auto filename for plot (not used if filenamecustom is supplied)
filenamecustom: optional filename for plot, otherwise default name is created
filepath: where to save image file, default=file.path("..//omopcept-plots")
canvas: some plot setups that override width,height,units "A4" "A4landscape" etc.
width: plot width, default=50
height: plot height, default=30
units: plot size units default='cm'
titlecolour: colour for main title, default='darkred'
backcolour: colour for background
graphtitle: optional title for graph, default NULL for none
graphsubtitle: optional subtitle for graph, default NULL for none
caption: optional text below plot, default=NULL
captiontxtsize: caption text size default=18,
captionjust: caption justification default="left",
captioncolour: caption text colour default="black",
messages: whether to print info messages, default=TRUE

Value

ggraph object

Examples

#showing how nodesizevar can be set from any column
bp <- omop_relations("Non-invasive blood pressure")
#> recursively querying concept relations of: Non-invasive blood pressure - may take more than a few seconds
#> step 1 of 1
#>    querying concept relations of: Non-invasive blood pressure - may take a few seconds
#> returning 8 concepts
nodesedges <- omop_graph_calc(bp)
nodesedges$nodes$testsizevar <- c(1:nrow(nodesedges$nodes))
omop_graph_vis(nodesedges, nodesizevar="testsizevar", nodesize = 5)
#> plot dims: w 50  h 30  u: cm 
#> saved graph file as omop_graph-graphopt-p1Dark2-legbottom3-nts9-nta0-ndomain_id-bwhite-egrey71-50x30cm.pdf