3 - Landscape graph construction and analysis with Graphab and graph4lg

UMR ThéMA - UBFC-CNRS, UMR Biogéosciences - UBFC-CNRS, ARP-Astrance



The rationale of graph4lg package in R is to make easier the construction and analysis of genetic and landscape graphs in landscape genetic studies (hence the name graph4lg, meaning Graphs for Landscape Genetics). This package provides users with tools for:

Each one of the included tutorials focuses on one of these points. This third tutorial will focus on landscape graph construction and analysis. It will describe the package functions allowing users to:

NOTA BENE: The package graph4lg integrates functions making possible the construction and analysis of landscape graphs but most of them are only wrappers that launch computations with Graphab software.

What is Graphab?

Graphab is an open source software tool dedicated to the modelling of landscape networks (Foltête, Clauzel, and Vuidel 2012). It was developed in ThéMA laboratory (Besançon, France) to make possible the creation of landscape graphs and to integrate a complete set of connectivity analysis functions in a single application.

Integrating Graphab functionalities into graph4lg aims at facilitating the import of Graphab output into R for subsequent analyses, including the comparison of genetic graphs and landscape graphs. All the functions calling Graphab software are described in graph4lg help files. However, users are invited to read the complete documentation manuals as well as other resources on Graphab project website. Some computations are not possible when using graph4lg but are possible when using directly Graphab software which can be downloaded for free on the same website. Foltête, Clauzel, and Vuidel (2012) paper also provides users with a synthetic description of the tool.

The function get_graphab from graph4lg package downloads automatically Graphab in the user’s machine, provided Java is installed on the machine. If it has already been downloaded, a message is displayed. Once Graphab has been downloaded in a machine, it is not necessary to download it again.


In the table below, we present all the functions from grapha4lg calling Graphab software:

Function Description
get_graphab Check if Graphab software (graphab-2.6.jar) is in the user’s machine and downloads it if necessary, provided Java is installed.
graphab_project Creates a Graphab project from a raster file (.tif)
graphab_link Creates a set of links (Euclidean or least cost paths, planar or complete)
graphab_graph Creates a graph from a set of links
graphab_metric Computes graph-theoretic metrics from a graph
graphab_modul Partition the graph into modules
graphab_pointset Imports a set of points in the project and relates them to habitat patches and corresponding metrics
get_graphab_metric Imports a table with the metrics computed and the patch properties
get_graphab_linkset Imports a table with the link properties for a given link set
graphab_to_igraph Imports a graph created with Graphab as an igraph graph object with node and link attributes

Data used in this tutorial

Here, we do not rely on genetic data sets and will only use Graphab projects already created for the vignettes.

Landscape graph construction

Project creation

When using Graphab, input data consists of an 8-bit raster file (.tif) with discrete cell values. One (or several) of these values will correspond to habitat cells on the raster. Contiguous habitat cells will form habitat patches (8-neighbourhood criterion), as soon as they exceed the minimum patch area when specified

This is the first step of habitat network modelling with Graphab. At this stage, a project is created. This project is given a name (proj_name), which will be the name of a directory containing geographical information layers relative to the project as well as a .xml file (“proj_name.xml”), which is the main project file.

The function graphab_project creates the project directory called proj_name either in the current working directory (default) or in the directory whose path is given as a proj_path argument. The function takes as arguments:

The two latter arguments are included in almost every function calling Graphab.

As an example, we will create a landscape graph from a simulated landscape raster with 5 land use categories (see map below).

load(file = paste0(system.file('extdata', package = 'graph4lg'), 
                               "/", "rast_simul50.RDa"))

r.spdf <- as(rast, "SpatialPixelsDataFrame")
r.df <- as.data.frame(r.spdf)

r.df$layer <- as.factor(r.df$rast_simul50)

g <- ggplot(r.df, aes(x=x, y=y)) + geom_tile(aes(fill = layer)) + coord_equal()+
  scale_fill_manual(values = c("#396D35", "#FB9013", "#EDC951", "#80C342", "black", "#396D35"),
                    labels = c("0 - Forest", "1 - Shrublands", "2 - Crops", 
                               "3 - Grasslands","4 - Artificial areas", "5 - Forest"),
                    name = "Land use type")+

Habitat patches will correspond to contiguous patches of raster cells equal to 0 or 5, as soon as they are larger than 200 hectares. They correspond to forests.

proj_name <- "graphab_example"

graphab_project(proj_name = proj_name,
                raster = "rast_simul50.tif",
                habitat = c(0, 5),
                minarea = 200)

Graph construction

We can now create a landscape graph with graphab_graph. This function takes as arguments:

proj_name is the only mandatory argument. Without any other argument, a non-thresholded graph is created for every link set present in the project. Names are created automatically.

In our example, we can create a planar graph with the following command:

graphab_graph(proj_name = proj_name,
              linkset = "lkst1",
              name = "graph")

Landscape graph analysis

Once the graph has been created, it can be analysed by computing graph-theoretic connectivity metrics or by partitioning its nodes.

Metric calculation

Many connectivity metrics can be computed from a landscape graph (see Baranyi et al. (2011) and Rayfield, Fortin, and Fall (2011) among other reviews on the subject). Graphab software includes a large range of connectivity metrics and its manual provides users with a comprehensive description of every one of them (see Graphab 2.6 manual).

The function graphab_metric computes these metrics. It takes as arguments:

Metrics fall into different categories. PC and IIC are global metrics and take only one value for the entire graph, which is returned in R environment when return_val=TRUE.

The other metrics are computed at the node level. When return_val=TRUE, a data.frame is returned in R environment specifying the value of the metric for every graph node.

The description of every metric is beyond the scope of this tutorial. We again invite users to read the help file (?graphab_metric) as well as the Graphab 2.6 user manual.

We will compute metrics on the graph "graph" in our example. First, we will compute the probability of connectivity at the global level. We set the dist and prob parameter such that: \(p(10km)=e^{-\alpha \times d_{ij}}=0.05\). We convert 10 km in cost-distance units for the computation.

# Global metric: PC
graphab_metric(proj_name = proj_name,
               graph = "graph",
               metric = "PC",
               dist = 10000,
               prob = 0.05,
               beta = 1,
               cost_conv = TRUE)
#> [1] "Project : graphab_example"    "Graph : graph"               
#> [3] "Metric : PC"                  "Dist : 10000"                
#> [5] "Prob : 0.05"                  "Beta : 1"                    
#> [7] "Value : 0.000903573300776615"

We obtain the value in an object of class list.

Now, using the same parameters, we compute the local metric Flux.

f <- graphab_metric(proj_name = proj_name,
               graph = "graph",
               metric = "F",
               dist = 10000,
               prob = 0.05,
               beta = 1,
               cost_conv = FALSE)
#> [[1]]
#> [1] "Project : graphab_example" "Graph : graph"            
#> [3] "Metric : F"                "Dist : 10000"             
#> [5] "Prob : 0.05"               "Beta : 1"

Every time a local metric is computed, it can be returned in R environment (here stored in f) but is also stored in the habitat patch attribute table, such that at the end this table contains every metric computed. We will see later that it can be very useful.

Graph partitioning

Another way to analyse landscape graph connectivity and topology is to make a partitioning. The principle is the same as that of modularity analyses of genetic graphs described in the second tutorial.

graphab_modul function carries out such an analysis, but before we perform some modifications in the current code, we recommend doing this analysis directly in Graphab software.

For the moment, it only creates a shapefile polygon layer with the Voronoi polygons corresponding to the groups of patches forming modules.

It takes as arguments:

For example:

graphab_modul(proj_name = proj_name,
              graph = "graph",
              dist = 10000,
              prob = 0.05,
              beta = 1)

NOTA BENE: All these analyses can be performed for several graphs created in the same project, provided their name is correctly indicated in the function arguments.

Using landscape graphs in landscape genetic analyses

Landscape graphs have been created and analysed by computing connectivity metrics and/or by partitioning them. The output of these analyses can be used in landscape genetic analyses, e.g. to assess the relationship between genetic variables measured in populations inhabiting the study landscape and the connectivity of its habitat patches. To do that, two approaches are here presented:

Point set import

Imagine that we have sampled several populations in the forests of the study landscape. We want to know in which landscape graph node lives every population, to get the corresponding connectivity metrics and compare them with the genetic response.

To do that, the function graphab_pointset imports a set of points into the Graphab project. It takes as arguments:

Point coordinates must be in the same coordinate reference system as the habitat patches (and initial raster layer). The function returns a data.frame with the properties of the nearest patch to every point in the point set, as well as the distance from each point to the nearest patch.

For example, we can add the point data.frame pts_pop_simul to the project with the following command.

# Point data frame
graphab_pointset(proj_name = proj_name,
                 linkset = "lkst1",
                 pointset = pts_pop_simul)

Note that when pointset is not the path to a shapefile point layer, such a layer will be created in the temp files of the user’s machine. It is therefore preferable to give the path to a shapefile layer as a pointset argument.

Landscape graph import

Finally, users can also create a graph from a Graphab link set and convert it into a graph object of class igraph. The imported graph has weighted links. Nodes attributes present in the Graphab project are included, including connectivity metrics when computed. A figure can be displayed automatically representing the graph on a map. The graph is given the topology of the selected link set.

The function graphab_to_igraph takes as arguments:

land_graph <- graphab_to_igraph(proj_name = proj_name,
                                linkset = "lkst1",
                                nodes = "patches",
                                weight = "cost",
                                fig = TRUE,
                                crds = TRUE)

crds_patches <- land_graph[[2]]
land_graph <- land_graph[[1]]

The function returns a list of two objects:

This graph can be plotted on a map with plot_graph_lg with node sizes proportional to habitat patch area and link width inversely proportional to cost-distances:

              crds = crds_patches,
              mode = "spatial",
              node_size = "Area")


We have seen in this third tutorial how to construct, analyse and import landscape graphs with Graphab and graph4lg. In the next and last tutorial, we will see how to compare genetic graphs and landscape graphs.


Baranyi, Gabriella, Santiago Saura, János Podani, and Ferenc Jordán. 2011. “Contribution of Habitat Patches to Network Connectivity: Redundancy and Uniqueness of Topological Indices.” Ecological Indicators 11 (5): 1301–10.
Foltête, Jean-Christophe, Céline Clauzel, and Gilles Vuidel. 2012. “A Software Tool Dedicated to the Modelling of Landscape Networks.” Environmental Modelling & Software 38: 316–27.
Rayfield, Bronwyn, Marie-Josée Fortin, and Andrew Fall. 2011. “Connectivity for Conservation: A Framework to Classify Network Measures.” Ecology 92 (4): 847–58.