In priviere/PPBstats: An R package to perform analysis found within PPB programmes

Unipart network for location analysis

This section deals with unipart network that represent relationship of germplasm diffusion between locations. The representation can be done for each germplasm or for each year.

Steps with `PPBstats`

Format the data with format_data_PPBstats()
get descriptive plot with plot()

Format the data

The format required is a data frame with the following compulsory columns as factor:

"location_parent" : the location associated to the parent seed lot
"location_child" : the location associated to the child seed lot
"relation_year_start" : the year when the relationship starts
"relation_year_end" : the year when the relationship stops

Possible column options are :

"germplasm_parent" : the germplasm associated to the parent seed lot
"year_parent" : the year of the last relationship of the parent seed lot
"germplasm_child" : the germplasm associated to the child seed lot
"year_child" : represents the year of the last relation event of the child seed lot

Other possibles column option are : "long_parent", "lat_parent", "long_child", "lat_child" to get map representation.

Note that data frame with unipart network for seed lots format can also be used.

The format of the data are checked by the function format_data_PPBstats() with the following arguments :

type : "data_network"
network_part : "unipart"
vertex_type : "location"
network_split : "germplasm" or "relation_year_start".

The function returns list of igraph object^[http://igraph.org/r/] coming from igraph::graph_from_data_frame().

data(data_network_unipart_sl)
head(data_network_unipart_sl)

Format and describe the data for each germplasm

For network_split = "germplasm", it returns a list with as many elements as germplam in the data as well as all germplasms merged in the first element of the list.

net_unipart_location_g = format_data_PPBstats(
  type = "data_network",
  data = data_network_unipart_sl, 
  network_part = "unipart", 
  vertex_type =  "location",
  network_split = "germplasm")

names(net_unipart_location_g)

The different representations are done with the plot() function.

For network representation, set plot_type = "network" diffusion event are display with curve. in_col can be settle to customize color of vertex. The curve between location represent the diffusion, the number of diffusion is displayed on a color scale.

p_net = plot(net_unipart_location_g, plot_type = "network", 
                          labels_on = "location", labels_size = 4)
names(p_net) # one element per germplasm, the first element with all the data
p_net$`germ-2`

Format and describe the data for each year

For network_split = "relation_year_start", it returns a list with as many elements as year in the data as well as all years merged in the first element of the list.

net_unipart_location_y = format_data_PPBstats(
  type = "data_network",
  data = data_network_unipart_sl,
  network_part = "unipart", 
  vertex_type =  "location",
  network_split = "relation_year_start")

names(net_unipart_location_y)

The different representations are done with the plot() function.

p_net = plot(net_unipart_location_y, plot_type = "network", 
                          labels_on = "location", labels_size = 4)
names(p_net) # one element per year, the first element with all the data
p_net$`2007-2008-2009`

With barplots, it represents the number of germplasm received or given.

p_bar = plot(net_unipart_location_y, plot_type = "barplot", x_axis = "location", in_col = "germplasm")
names(p_bar) # one element per year, the first element with all the data
p_bar = p_bar$`2007-2008-2009`
p_bar$barplot$received
p_bar$barplot$given

Location present on the network can be displayed on a map with plot_type = "map". When using map, do not forget to use credit : Map tiles by Stamen Design, under CC BY 3.0. Data by OpenStreetMap, under ODbL.

p_map = PPBstats:::plot.data_network(net_unipart_location_y[1], plot_type = "map", labels_on = "location")
# Note if you want to do it on all element of the list, you should use 
# plot(net_unipart_location_y, plot_type = "map", labels_on = "location")
# Here we use PPBstats:::plot.data_network only not to ask to often the map server that may bug if there are too many query
# We use ::: because the function is not exported as it is an S3 method
p_map$`2007-2008-2009`

As well as plot information regarding a variable on map with a pie with plot_type = "map" and by setting arguments data_to_pie and vec_variables:

# y1 is a quantitative variable
p_map_pies_y1 = PPBstats:::plot.data_network(net_unipart_location_y[1], data_to_pie, plot_type = "map", vec_variables = "y1")
p_map_pies_y1$`2007-2008-2009`

# y2 is a qualitative variable
p_map_pies_y2 = PPBstats:::plot.data_network(net_unipart_location_y[1], data_to_pie, plot_type = "map", vec_variables = "y2")
p_map_pies_y2$`2007-2008-2009`