Movatterモバイル変換

A graph object of classigraph.

A character string indicating the nature of theinput data from which come the attributes to add to the nodes.

• If 'input = "shp"', then attributes come from the attribute table ofa shapefile layer of type point.

• If 'input = "df"', then attributes come from an object of classdata.frame

In both cases, input attribute table or dataframe must have a column withthe exact same values as the node IDs.

(only if 'input = "df"') The name of the object ofclassdata.frame with the attributes to add to the nodes.

(only if 'input = "shp"') The path (character string) to thedirectory containing the shapefile layer of type point whose attributetable contains the attributes to add to the nodes.

(only if 'input = "shp"') The name (character string) of theshapefile layer of type point (without extension, ex.: "nodes" refersto "nodes.shp" layer) whose attribute table contains the attributesto add to the nodes.

The name (character string) of the column with the nodes namesin the input data (column of the attribute table or of the dataframe).

A character string (vector) indicating which columns of theinput data will be added as nodes' attributes.By default, 'include = "all"', i.e. every column of the input data is added.Alternatively, 'include' can be a vector with the names of the columns to add(ex.: "c('x', 'y', 'pop_name')").

The path to the project .xml file.

An object of classdata.frame with at least 4 columnsof data used to perform the test.4 columns must be called "n1", "n2", "r1" and "r2".

• n1 and n2 are the sizes of the samples from which r1 and r2were computed respectively.

• r1 and r2 are Pearson's correlation coefficients

An object of classigraph. Its nodes must have names.

A character string indicating the algorithm used to createthe modules withigraph.

• Ifalgo = 'fast_greedy' (default),functioncluster_fast_greedy fromigraphis used (Clauset et al., 2004).

• Ifalgo = 'walktrap', functioncluster_walktrapfromigraph is used (Pons et Latapy, 2006) with 4 steps(default options).

• Ifalgo = 'louvain', functioncluster_louvainfromigraph is used (Blondel et al., 2008). In that case, the numberof modules created in each graph is imposed.

• Ifalgo = 'optimal', functioncluster_optimalfromigraph is used (Brandes et al., 2008) (can be very long).In that case, the number of modules created in each graph is imposed.

(optional, default = NULL) A character string indicatingwhether the links of the graph are weighted by distances or by similarityindices. It is only used to compute the modularity index. It can be:

• 'distance': Link weights correspond to distances. Nodes that are closeto each other will more likely be in the same module.

• 'similarity': Link weights correspond to similarity indices. Nodes thatare similar to each other will more likely be in the same module. Inverselink weights are then used to compute the modularity index.

(optional , default = NULL) A numeric or integer valueindicating the number of modules in the graph. When this number is notspecified, the optimal value is retained.

An object of classigraph. Its nodes must have names.

Character vector specifying the graph-theoreticmetrics computed at the node-level in the graphsGraph-theoretic metrics can be:

• Degree (metrics = c("deg", ...))

• Closeness centrality index (metrics = c("close",...))

• Betweenness centrality index (metrics = c("btw",...))

• Strength (sum of the weights of the links connected to a node)(metrics = c("str",...))

• Sum of the inverse weights of the links connected to anode (metrics = c("siw", ...), default)

• Mean of the inverse weights of the links connected to anode (metrics = c("miw", ...))

By default, the vectormetrics includes all these metrics.

Logical which indicates whether the links are weighted duringthe calculation of the centrality indices betweenness and closeness.(default:weight = TRUE). Link weights are interpreted as distanceswhen computing the shortest paths. They should then be inversely proportionalto the strength of the relationship between nodes (e.g. to fluxes).

A symmetricmatrix ordist object withpairwise geographical Euclidean distances between populations or samplesites. It will be the explanatory variable, and only values from the offdiagonal lower triangle will be used.

A symmetricmatrix ordist object with pairwiselandscape distances between populations or sample sites. These distances canbe cost-distances or resistance distances, among others. It will be theexplained variable, and only values from the off diagonal lower trianglewill be used.

A numeric value or numeric vector with Euclidean distancesto convert into cost-distances.

A character string indicating the method used to fit the model.

• If 'method = "log-log"' (default), then the model takes thefollowing form : log(ld) ~ A + B * log(euc)

• If 'method = "lm"', then the model takes the following form :ld ~ A + B * euc

Logical (default = TRUE) indicating whether a figure is plotted

(if 'fig = TRUE') Character string indicating the colorused to plot the line (default: "blue"). It must be a hexadecimal colorcode or a color used by default in R.

(if 'fig = TRUE') Character string indicating the colorused to plot the points (default: "#999999"). It must be a hexadecimal colorcode or a color used by default in R.

A coordinate in degrees

An object of classdata.frame

A character string indicating the name of the column with the IDof the origins

A character string indicating the name of the column with the IDof the arrivals

A character string indicating the name of the column with thevalues corresponding to each pair

Point 1 longitude in radians

Point 1 latitude in radians

Point 2 longitude in radians

Point 2 latitude in radians

Point 1 longitude in radians

Point 1 latitude in radians

Point 2 longitude in radians

Point 2 latitude in radians

Point 1 longitude in radians

Point 1 latitude in radians

Point 2 longitude in radians

Point 2 latitude in radians

project name, project dir in which proj_name.xml is found

raster.tif INT2S path or present in wd,

habitat code in the raster file

default 0, minimum habitat size in ha

default NULL nodata code in the raster file

A symmetricmatrix ordist object with pairwisegenetic distances between populations or sample sites.

A symmetricmatrix ordist object with pairwiselandscape distances between populations or sample sites. These distancescan be Euclidean distances, cost-distances or resistance distances,among others.

A numeric or integer value indicating the interval betweenthe different distance thresholds for which the correlation coefficientsare computed.

(optional) The minimum distance threshold value at which thecorrelation coefficient is computed.

(optional) The maximum distance threshold value at which thecorrelation coefficient is computed.

Logical (default = TRUE) indicating whether a figure is plotted.

(optional) A numeric or integer value used to removegenetic distance values from the data before the calculation.All genetic distances values above 'thr_gd' are removed from the data.This parameter can be used especially when there are outliers.

(optional, if fig = TRUE) A character string indicating thecolor used to plot the line (default: "blue"). It must be a hexadecimal colorcode or a color used by default in R.

(optional, if fig = TRUE) A character string indicating thecolor used to plot the points (default: "#999999"). It must be a hexadecimalcolor code or a color used by default in R.

A symmetricmatrix or adist object with pairwisedistances between nodes

The number of classes to create to search for thethreshold value without testing all the possibilities. By default,'val_step = 20'.

An object of classgenind that contains the multilocusgenotype (format 'locus') of the individuals as well as their populationand their geographical coordinates.

A character string indicating the method used to compute themultilocus genetic distance between populations

• If 'dist = 'basic” (default), then the multilocus genetic distance iscomputed using a Euclidean genetic distance formula (Excoffier et al., 1992)

• If 'dist = 'weight”, then the multilocus genetic distance is computedas in Fortuna et al. (2009). It is a Euclidean genetic distance giving moreweight to rare alleles

• If 'dist = 'PG”, then the multilocus genetic distance is computed asin popgraph::popgraph function, following several steps of PCA and SVD(Dyer et Nason, 2004).

• If 'dist = 'PCA”, then the genetic distance is computed following aPCA of the matrix of allelic frequencies by population. It is a Euclideangenetic distance between populations in the multidimensional space definedby all the independent principal components.

A character string indicating the formula used to compute thecovariance matrix from the distance matrix

• If 'cov = 'sq” (default), then the covariance matrix is calculatedfrom the matrix of squared distances as in Everitt et Hothorn (2011)

• If 'cov = 'dist”, then the covariance matrix is calculated from thematrix of distances as in Dyer et Nason (2004) and popgraph function

A character string indicating the way the partial correlationmatrix is computed from the covariance matrix.

• If 'pcor = 'magwene”, the steps followed are the same as inMagwene (2001) and in popgraph::popgraph function. It is the recommendedoption as it meets mathematical requirements.

• If 'pcor = 'other”, the steps followed are the same as usedby Fortuna et al. (2009). They are not consistent with the approachof Magwene (2001).

A numeric value corresponding to the statistical tolerancethreshold used to test the difference from 0 of the partial correlationcoefficients. By default, 'alpha=0.05'.

A character string indicating the method used to test thesignificance of the partial correlation coefficients.

• If 'test = 'EED” (default), then the Edge Exclusion Deviancecriterion is used (Whittaker, 2009). Although other methods exist, this isthe most common and thus the only one implemented here.

A character string indicating the way of adjusting p-values toassess the significance of the p-values

• If 'adj = 'none” (default), there is no p-value adjustment correction

• If 'adj = 'holm”, p-values are adjusted using the sequentialBonferroni correction (Holm, 1979)

• If 'adj = 'bonferroni”, p-values are adjusted using the classicBonferroni correction

• If 'adj = 'BH”, p-values are adjusted using Benjamini et Hochberg(1995) correction controlling false discovery rate

A character string indicating the matrices included inthe output list.

• If 'output = 'all” (default), then D (distance matrix),C (covariance matrix), Rho (partial correlation matrix),M (graph incidence matrix) and S (strength matrix) are included

• If 'output = 'dist_graph”, then the distance matrix D is returnedonly with the values corresponding to the graph edges

• If 'output = 'str_graph”, then the strength values matrix S isreturned only with the values corresponding to the graph edges

• If 'output = 'inc”, then the binary adjacency matrix M is returned

• If 'output = 'igraph”, then a graph of classigraphis returned

A symmetric (pairwise)matrix or adist objectwhose elements will be the links' weights

(optional) A symmetric (pairwise) distancematrixor adist object whose values will be used for the pruning basedon the threshold value.

The threshold value (logically between min(mat_thr)and max(mat_thr))(integer or numeric)

• If 'mode = 'larger” (default), all the links whose weight is largerthan 'thr' are removed.

• If 'mode = 'lower”, all the links whose weight is lowerthan 'thr' are removed.

A symmetric (pairwise)matrix or adist objectwhose elements will be the links' weights

(optional) A symmetric (pairwise) distancematrixor adist object whose values will be used for the pruning method.

Which topology does the created graph have?

• If 'topo = 'gabriel” (default), the resulting graph will be aGabriel graph (Gabriel et al., 1969). It means that there is a linkbetween nodes x and y if and only ifd_{xy}^{2} \leq \min(\sqrt{d_{xz}^{2}+d_{yz}^{2}}),with z any other node of the graph.

• If 'topo = 'mst”, the resulting graph will have the topologyof a minimum spanning tree. It means that the graph will not includeany cycle (tree) and it will be the subgraph with a tree topology withthe minimum total links' weight (based on 'mat_topo' values).

• If 'topo = 'percol”, if the link of the resulting graph with theminimum weight is removed, then the graph breaks into two components.

• If 'topo = 'comp”, a complete graph whose links are weighted withvalues from 'mat_w' is created.

• If 'topo = 'knn”, a k-nearest neighbor graph whose links areweighted with values from 'mat_w' is created. If the distance between node iand node j is among the k-th smallest distances between node i and the othernodes according to distances in matrix 'mat_topo', then there is a linkbetween i and j in the resulting graph. Therefore, a node can be connectedto more than two nodes because the nearest node to node j is not necessarilyamong the k nearest neighbors to node i. Let d1 be the smallest distancefrom node i to other nodes, if there are k nodes or more at this distancefrom node i, they are all connected to i. If there are less than k nodesconnected to i at a distance d1, then we consider nodes at a distance d2from i. In the latter case, all the nodes at a distance d2 from i areconnected to i.

(if 'topo = 'knn”) An integer which indicates the number ofnearest neighbors considered to create the K-nearest neighbor graph. k mustbe lower than the total number of nodes minus 1.

A character string with the path leading to the GENEPOP filein format .txt, or alternatively the name of this file in the workingdirectory.

The number of loci in the GENEPOP file (integer or numeric).

(optional) Populations' names in the same orderas in the GENEPOP file.Vector object (class character) of the same length as the numberof populations.Without this parameter, populations are numbered from 1 to the numberof populations.

Number indicating whether alleles are codedwith 3 (default) or 2 digits.

An object of classgenindfrom packageadegenet.

A character string indicating the option used to select whatthe function will return:

• Ifoutput = "data.frame"(default), then the function willreturn an object 'x' of classdata.frame ready to be saved as atext file with the following command:write.table(x, file = "file_name.txt", quote=FALSE,row.names=FALSE, col.names=FALSE)

• Ifoutput = "path_to_file/file_name.txt", then the functionwill write a text file named 'file_name.txt' in the directory correspondingto 'path_to_file'. Without 'path_to_file', the text file is written in thecurrent working directory. The text file has the format required by GENEPOPsoftware.

An object of classgenindfrom packageadegenet.

A character string of the formoutput = "path_to_file/file_name.txt". Then, the functionwill write a text file named 'file_name.txt' in the directory correspondingto 'path_to_file'. Without 'path_to_file', the text file is written in thecurrent working directory. The text file has the format required by STRUCTUREsoftware.

Logical indicating whether a message says if Graphab has beendownloaded or not.

Logical indicating whether the function returns a 1 or a 0to indicate if Graphab has been downloaded or not.

A character string indicating the Graphab project name.The project name is also the name of the project directory in which thefile proj_name.xml is.

A character string indicating the name of the link setwhose properties are imported. The link set has been created with Graphabor usinggraphab_link function.

(optional) A character string indicating the path to thedirectory that contains the project directory. It should be used when theproject directory is not in the current working directory. Default is NULL.When 'proj_path = NULL', the project directory is equal togetwd().

A character string indicating the Graphab project name.The project name is also the name of the project directory in which thefile proj_name.xml will be created.

(optional, default=NULL) A character string indicating thename of the link set used to create the graph. Link sets can be createdwithgraphab_link.

(optional) A character string indicating the path to thedirectory that contains the project directory. It should be used when theproject directory is not in the current working directory. Default is NULL.When 'proj_path = NULL', the project directory is equal togetwd().

A character string indicating the Graphab project name.The project name is also the name of the project directory in which thefile proj_name.xml is.

(optional) A character string indicating the path to thedirectory that contains the project directory. It should be used when theproject directory is not in the current working directory. Default is NULL.When 'proj_path = NULL', the project directory is equal togetwd().

A character string indicating the Graphab project name.The project name is also the name of the project directory in which thefile proj_name.xml will be created.

A character string equal to either 'all' (default) or 'habitat'indicating whether the returned codes are all the codes of the source rasterused for creating the project or only the code corresponding to thehabitat patches.

(optional) A character string indicating the path to thedirectory that contains the project directory. It should be used when theproject directory is not in the current working directory. Default is NULL.When 'proj_path = NULL', the project directory is equal togetwd().

A numeric vector with positive values

A logical value indicating whether the computed coefficientis biased or not. Unbiased value are equal to n/(n-1) times the biased ones.

The first graph object

• Ifmode = 'graph' (default),x is a graph object ofclassigraph.Then, its nodes must have the same names as in graphy.

• Ifmode = 'data.frame',x refers to a column ofthedata.frame 'data'.Thenx must be a character string indicating the name of thecolumn of 'data' with the modules' labels of the nodes in the first graph.In that case, the column can be of classnumeric,characterorfactor but will be converted into anumeric vectorin any case.

• Ifmode = 'vector',x is a vector ofclasscharacter,factor ornumeric.In that case, it must have the same length as vectory andwill be converted into anumeric vector.

The second graph objectSame classes possible as forx. Must be of the same format asx

A character string indicating whether x and y are igraph objects,vectors or columns from a data.frame.mode can be 'graph','data.frame' or 'vector'.

(if x and y are igraph objects) A numeric or integer valueor a numeric vector with 2 elements indicating the number of modules tocreate in both graphs.

• Ifnb_modul is a numeric value, then the same number of modulesare created in both graphs.

• Ifnb_modul is a numeric vector of length 2, then thenumbers of modules created in graphsx andy are thefirst and second elements ofnb_modul, respectively.

(if x and y are igraph objects) A character string indicating thealgorithm used to create the modules withigraph.

• Ifalgo = 'fast_greedy' (default),functioncluster_fast_greedyfromigraph is used (Clauset et al., 2004).

• Ifalgo = 'walktrap' (default), functioncluster_walktrapfromigraph is used (Pons et Latapy, 2006) with4 steps (default options).

• Ifalgo = 'louvain', functioncluster_louvainfromigraph is used (Blondel et al., 2008).In that case, the number of modules created in each graph is imposed.

• Ifalgo = 'optimal', functioncluster_optimalfromigraph is used (Brandes et al., 2008) (can be very long).In that case, the number of modules created in each graph is imposed.

(optional, if x and y are igraph objects,default is 'none') A character string indicating whether the links of thegraph are weighted by distances or by similarity indices. It is only usedto compute the modularity index. It can be:

• 'distance': Link weights correspond to distances. Nodes that are closeto each other will more likely be in the same module.

• 'similarity': Link weights correspond to similarity indices. Nodes thatare similar to each other will more likely be in the same module. Inverselink weights are then used to compute the modularity index.

• 'none': Links are not weighted for the computation, which is onlybased on graph topology.

Two different weightings can be used to create the modules of the two graphs.

• Ifnode_inter is a character string, then the same linkweighting is used for both graphs.

• Ifnode_inter is a character vector of length 2, thenthe link weighting used by the algorithm to create the modules ofgraphsx andy is determined by the first and second elementsofnode_inter, respectively.

(if x and y are columns from a data.frame) An object of classdata.frame with at least two columns and as many rows as there are nodesin the graphs compared. The columns indicate the modules of each node in2 different classifications.

An object of classigraph.Its nodes must have the same names as in graphy.

An object of classigraph.Its nodes must have the same names as in graphx.

Two-element character vector specifying the graph-theoreticmetrics computed at the node-level in the graphs or the node attributevalues to be correlated to these metrics.Graph-theoretic metrics can be:

• Degree (metrics = c("deg", ...))

• Closeness centrality index (metrics = c("close",...))

• Betweenness centrality index (metrics = c("btw",...))

• Strength (sum of the weights of the links connected to a node)(metrics = c("str",...))

• Sum of the inverse weights of the links connected to anode (metrics = c("siw", ...), default)

• Mean of the inverse weights of the links connected to anode (metrics = c("miw", ...))

Node attributes must have the same names as in theigraph object,and must refer to an attribute with numerical values.The vectormetrics is composed of two character values.When a node attribute has the same name as a metric computable from thegraph, node attributes are given priority.

A character string indicating which correlation coefficientis to be computed ("pearson","kendall" or"spearman" (default)).

Logical which indicates whether the links are weighted duringthe calculation of the centrality indices betweenness and closeness.(default:weight = TRUE). Link weights are interpreted as distanceswhen computing the shortest paths. They should then be inversely proportionalto the strength of the relationship between nodes (e.g. to fluxes).

Logical. Should significance testing be performed?(default = TRUE)

Adata.frame with the spatialcoordinates of the point set (the graph nodes). It must have three columns:

• ID: A character string indicating the name of the points(graph nodes).

• x: A numeric or integer indicating the longitude of the graph nodes.

• y: A numeric or integer indicating the latitude of the graph nodes.

A character string indicating the name of the columnofcrds with the point IDs

A character string indicating the name of the columnofcrds with the point longitude

A character string indicating the name of the columnofcrds with the point latitude

A character string indicating whether the links ofthe graph are weighted by Euclidean distances (TRUE)(default) or not (FALSE).When the graph links do not have weights in Euclidean distances, each linkis given a weight of 1.

A graph object of classigraph.Its nodes must have the same names as in graphy.

A graph object of classigraph.Its nodes must have the same names as in graphx.

Adata.frame with the spatialcoordinates of the graph nodes (bothx andy).It must have three columns:

• ID: Name of the graph nodes (character string).The names must be the same as the node names of the graphs ofclassigraph (igraph::V(graph)$name)

• x: Longitude of the graph nodes (numeric or integer).

• y: Latitude of the graph nodes (numeric or integer).

A graph object of classigraph

Logical. If TRUE (default), then the column 'link' of theoutput data.frame contains the weights of the links. If FALSE,it contains only 0 and 1.

A graph object of classigraph

(if 'mode = 'spatial”) Adata.frame with the spatialcoordinates of the graph nodes. It must have three columns:

• ID: Name of the graph nodes (will be converted into character string).The names must the same as the node names of the graph object ofclassigraph (igraph::V(graph)$name)

• x: Longitude (numeric or integer) of the graph nodes in the coordinatesreference system indicated with the argument crds_crs.

• y: Latitude (numeric or integer) of the graph nodes in the coordinatesreference system indicated with the argument crds_crs.

Indicates which shapefile layers will be created

• If 'mode = 'both” (default), then two shapefile layers are created,one for the nodes and another for the links.

• If 'mode = 'node”, a shapefile layer is created for the nodes only.

• If 'mode = 'link”, a shapefile layer is created for the links only.

An integer indicating the EPSG code of the coordinatesreference system to use.The projection and datum are given in the PROJ.4 format.

A character string indicating the suffix of the name ofthe layers to be created.

A character string corresponding to the path to the directoryin which the shapefile layers will be exported. Ifdir_path = "wd",then the layers are created in the current working directory.

(not considered if 'mode = 'link”) Logical. Should graphnode attributes integrated in the attribute table of the node shapefilelayer? (default: FALSE)

A graph object of classigraph with n nodes.It is the observed graph thatpred_graph is supposed to approach.

A graph object of classigraph with n nodes.It is the predicted graph that is supposed to be akin toobs_graph.

A character string specifying which index to compute in orderto compare the topologies of the graphs.

• If 'mode = 'mcc” (default), the Matthews CorrelationCoefficient (MCC) is computed.

• If 'mode = 'kappa”, the Kappa index is computed.

• If 'mode = 'fdr”, the False Discovery Rate (FDR) is computed.

• If 'mode = 'acc”, the Accuracy is computed.

• If 'mode = 'sens”, the Sensitivity is computed.

• If 'mode = 'spec”, the Specificity is computed.

• If 'mode = 'prec”, the Precision is computed.

Logical (TRUE or FALSE) specifying whether both graphsare directed or not.

A character string indicating the Graphab project name.The project name is also the name of the project directory in which thefile proj_name.xml is. It can be created withgraphab_project

A character string indicating the way capacities arecomputed. It must be either:

• mode='area'(default): The capacity of the patches is computedas the area of each habitat patch. The argumentexp makes itpossible to raise area to a power given by an exposant.

• mode='ext_file': The capacity of the patches is given by anexternal .csv file. See argumentext_file below.

• mode='neigh': The capacity is computed depending on theneighbouring raster cells from each habitat patch. The number of cellswith a value given bycodes argument is summed up to thedistancethr. This number can be weighted according to theweight argument.

(optional, default=NULL) An integer value or vectorspecifying the codes corresponding to the habitat pixel whose correspondingpatches are included to compute the capacity as the area of the habitatwhenmode='area'. Patches corresponding to other initial habitatcodes are weighted by 0.

An integer value specifying the power to which patch area areraised whenmode='area'. When not specified,exp=1 by default.

A character string specifying the name of the .csv file inwhich patch capacities are stored. It must be located either in the workingdirectory or in the directory defined byproj_path. It must haveas many rows as there are patches in the project and its column namesmust include 'Id' and 'Capacity'. The 'Id' column must correspond to thepatch ID in the 'patches' layer (seeget_graphab_metric).The 'Capacity' column must contain the corresponding patch capacities toassign each patch.

(optional, default=NULL) An integer or numeric value indicatingthe maximum distance in cost distance units (except whencost_conv = TRUE) at which cells are considered for computing thecapacity whenmode='neigh'.

(optional, default=NULL) A character string indicating thename of the link set used to take distance into account when computingthe capacity. Only used whenmode='neigh'. Link sets can becreated withgraphab_link.

An integer value or a vector of integer values specifying thecodes of the raster cells taken into account when computing the capacity inthe neighbourhood of the patches, whenmode='neigh'.

FALSE (default) or TRUE. Logical indicating whether numericthr values are converted from cost-distance into Euclidean distanceusing a log-log linear regression. See alsoconvert_cdfunction. Only used whenmode='neigh'.

A logical indicating whether the cells are weighted by aweight decreasing with the distance from the patches (TRUE) or not (FALSE).The weights follow a negative exponential decline such thatwi = exp(-alpha*di), where wi is the weight of cell i, di its distance fromthe patch and alpha a parameter determined such that wi = 0.05 when di = thr.

(optional) A character string indicating the path to thedirectory that contains the project directory. It should be used when theproject directory is not in the current working directory. Default is NULL.When 'proj_path = NULL', the project directory is equal togetwd().

(optional, default = NULL) Integer or numeric valueindicating RAM gigabytes allocated to the java process. Increasing thisvalue can speed up the computations. Too large values may not be compatiblewith your machine settings.

A character string indicating the Graphab project name.The project name is also the name of the project directory in which thefile proj_name.xml is. It can be created withgraphab_project

A character string indicating the name of the graph with thelinks from which the corridors are computed.This graph has been created with Graphab or usinggraphab_graphfunction and is associated with a link set.Only the links present in the graph will be used in the computation.

An integer or numeric value indicating the maximum costdistance from the least-cost paths considered for creating the corridors,in cost distance units (except whencost_conv = TRUE).

(optional, default = "raster") A character string indicatingwhether the output is a raster file or a shapefile layer.

FALSE (default) or TRUE. Logical indicating whether numericthr values are converted from cost-distance into Euclidean distanceusing a log-log linear regression. See alsoconvert_cdfunction. Only used whenmode='neigh'.

(optional) A character string indicating the path to thedirectory that contains the project directory. It should be used when theproject directory is not in the current working directory. Default is NULL.When 'proj_path = NULL', the project directory is equal togetwd().

(optional, default = NULL) Integer or numeric valueindicating RAM gigabytes allocated to the java process. Increasing thisvalue can speed up the computations. Too large values may not be compatiblewith your machine settings.

A character string indicating the Graphab project name.The project name is also the name of the project directory in which thefile proj_name.xml is. It can be created withgraphab_project

(optional, default=NULL) A character string indicating thename of the link set used to create the graph. Iflinkset=NULL, everylink set present in the project will be used to create a graph. Link setscan be created withgraphab_link.

(optional, default=NULL) A character string indicating thename of the graph created. Ifname=NULL, a name will be created. Ifbothlinkset=NULL andname=NULL, then a graph will be createdfor every link set present in the project and a name will be created everytime. In the latter case, a unique name cannot be specified. Link setscan be created withgraphab_link.

(optional, default=NULL) An integer or numeric value indicatingthe maximum distance associated with the links of the created graph. Itallows users to create a pruned graph based on a distance threshold. Note thatwhen the link set used has a planar topology, the graph is necessarily apruned graph (not complete) and adding this threshold parameter can removeother links. When the link set has been created with cost-distances, theparameter is expressed in cost-distance units whereas when the link set isbased upon Euclidean distances, the parameter is expressed in meters.

FALSE (default) or TRUE. Logical indicating whether numericthr values are converted from cost-distance into Euclidean distanceusing a log-log linear regression. See alsoconvert_cdfunction.

(optional) A character string indicating the path to thedirectory that contains the project directory. It should be used when theproject directory is not in the current working directory. Default is NULL.When 'proj_path = NULL', the project directory is equal togetwd().

(optional, default = NULL) Integer or numeric valueindicating RAM gigabytes allocated to the java process. Increasing thisvalue can speed up the computations. Too large values may not be compatiblewith your machine settings.

A character string indicating the Graphab project name.The project name is also the name of the project directory in which thefile proj_name.xml is. It can be created withgraphab_project

A character string indicating the name of the raster to becreated after the interpolation.

An integer indicating the spatial resolution in meters of theraster resulting from the metric interpolation.

A character string indicating the name of the link set usedfor the interpolation. It should be the one used to create the used graphand the metric.

A character string indicating the name of the graph from whichthe metric was computed and whose links are considered for a potentialmulti-linkage with patches.This graph has been created with Graphab or usinggraphab_graphfunction and is associated with a link set.

A character string indicating the name of the already computedmetric to be interpolated.

A numeric or integer value specifying the distance at which weassume a probability equal toprob during the interpolation.It is used to set\alpha for computing probabilities associatedwith distances between each pixel and the neighboring patch(es) such thatprobability between patch i and pixel j isp_{ij}= e^{-\alpha d_{ij}}.

A numeric or integer value specifying the probabilityat distancedist. By default,code=0.05. It is used to set\alpha (see paramdist above).

(default NULL) If NULL, the value of each pixel is computed fromthe value of the metric at the nearest habitat patch, weighted by aprobability depending on distance. If an integer, the value of each pixeldepends on the values of the metric taken at several of the nearest habitatpatches, up to a distance (cost or Euclidean distance, depending on the typeof linkset) equal tothr.

Logical (default = FALSE) only used ifthr is not NULL,and specifying whether multiple values are summed up (TRUE) or averagedafter being weighted by probabilities.

(optional) A character string indicating the path to thedirectory that contains the project directory. It should be used when theproject directory is not in the current working directory. Default is NULL.When 'proj_path = NULL', the project directory is equal togetwd().

(optional, default = NULL) Integer or numeric valueindicating RAM gigabytes allocated to the java process. Increasing thisvalue can speed up the computations. Too large values may not be compatiblewith your machine settings.

A character string indicating the Graphab project name.The project name is also the name of the project directory in which thefile proj_name.xml is. It can be created withgraphab_project

A character string indicating whether links between patchesare computed based on:

• Shortest cost distances:distance='cost' (default)

• Straight Euclidean distances:distance='euclid'

In the resulting link set, each link will be associated with itscorresponding cost-distance and the length of the least-cost path in meters(ifdistance='cost') or with its length in Euclidean distance(ifdistance='euclid')

A character string indicating the name of the created linkset.

This argument could be:

• Adata.frame indicating the cost values associated to eachraster cell value. These values refer to the raster used to create theproject withgraphab_project. The data.frame must have twocolumns:

  • 'code': raster cell values

  • 'cost': corresponding cost values

• The path to an external raster file in .tif format with cost values.

A character string indicating the topology of the createdlink set. It can be:

• Planar (topo='planar' (default)): a planar set of links iscreated. It speeds up the computation but will prevent from creatingcomplete graphs withgraphab_graph.

• Complete (topo='complete'): a complete set of links is created.A link is computed between every pair of patches.

(optional, default = FALSE) A logical indicating whetherlinks crossing patches are removed (TRUE).

(optional) A character string indicating the path to thedirectory that contains the project directory. It should be used when theproject directory is not in the current working directory. Default is NULL.When 'proj_path = NULL', the project directory is equal togetwd().

(optional, default = NULL) Integer or numeric valueindicating RAM gigabytes allocated to the java process. Increasing thisvalue can speed up the computations. Too large values may not be compatiblewith your machine settings.

A character string indicating the Graphab project name.The project name is also the name of the project directory in which thefile proj_name.xml is.

A character string indicating the name of the graph on whichthe metric is computed. This graph has been created with Graphabor usinggraphab_graph function and is associatedwith a link set. Only the links present in the graph and their correspondingweights will be used in the computation, together with patch areas.

A character string indicating the metric which will be computedon the graph. This metric can be:

• A global metric:

  • Probability of Connectivity (metric = 'PC'): Sum of products ofarea of all pairs of patches weighted by their interaction probability,divided by the square of the area of the study zone.This ratio is the equivalent to the probability that two points randomlyplaced in the study area are connected.

  • Equivalent Connectivity (metric = 'EC'): Square root of thesum of products of capacity of all pairs of patches weighted by theirinteraction probability. This is the size of a single patch (maximallyconnected) that would provide the same probability of connectivity as theactual habitat pattern in the landscape (Saura et al., 2011).

  • Integral Index of Connectivity (metric = 'IIC'): For theentire graph: product of patch areas divided by the number of linksbetween them, the sum is divided by the square of the area of the studyzone. IIC is built like the PC index but using the inverse of a topologicaldistance rather than a negative exponential function of the distancebased on the link weight.

• A local metric:

  • Flux (metric = 'F'): For the focal patch i : sum of areaof patches other than i and weighted according to their minimum distanceto the focal patch through the graph. This sum is an indicator of thepotential dispersion from the patch i or, conversely to the patch i

  • Betweenness Centrality index (metric = 'BC'): Sum of theshortest paths through the focal patch i, each path is weighted by theproduct of the areas of the patches connected and of their interactionprobability. All possible paths between every pair of patches isconsidered in this computation.

  • Interaction Flux (metric = 'IF'): Sum of products of the focalpatch area with all the other patches, weighted by their interactionprobability.

  • Degree (metric = 'Dg'): Number of edges connected to thenode i i.e. number of patches connected directly to the patch i.

  • Closeness Centrality index (metric = 'CCe'): Mean distancefrom the patch i to all other patches of its component k.

  • Current Flux (metric = 'CF'): Sum of currents passing throughthe patch i.c_{i}^{j} represents the current through the patch i whencurrents are sent from all patches (except j) to the patch j.The patch j is connected to the ground.

• A delta metric:

  • delta Probability of Connectivity (metric = 'dPC'): Rate ofvariation between the value of PC index and the value of PC' correspondingto the removal of the patch i. The value ofdPC is decomposedinto three parts:

    • dPC_{area} is the variation induced by the area lost after removal;

    • dPC_{flux} is the variation induced by the loss of interactionbetween the patch i and other patches;

    • dPC_{connector} is the variation induced by the modification ofpaths connecting other patches and initially routed through i.

For most metrics, the interaction probability is computed for each pair ofpatches from the path that minimizes the distance d (or the cost) betweenthem. It then maximizes{e}^{-\alpha d_{ij}} for patches i and j.To use patch capacity values different from the patch area, please usedirectly Graphab software.

A logical (default = FALSE) indicating whether the'multihabitat' mode is used when computing the metric. It only applies tothe following metrics: 'EC', 'F', 'IF' and 'BC'. If TRUE, then the projectmust have been created with the optionnomerge=TRUE. It then returnsseveral columns with metric values including the decomposition of thecomputation according to the type of habitat of every patch.Be careful, this option is in development and we cannot guarantee theresults are correct.

A numeric or integer value specifying the distance at whichdispersal probability is equal toprob. This argument is mandatoryfor weighted metrics (PC, F, IF, BC, dPC, CCe, CF) but not used for others.It is used to set\alpha for computing dispersal probabilities associatedwith all inter-patch distances such that dispersal probability betweenpatches i and j isp_{ij}= e^{-\alpha d_{ij}}.

A numeric or integer value specifying the dispersal probabilityat distancedist. By default,code=0.05. It is used to set\alpha (see paramdist above).

A numeric or integer value between 0 and 1 specifying theexponent associated with patch areas in the computation of metricsweighted by patch area. By default,beta=1. Whenbeta=0, patchareas do not have any influence in the computation.

FALSE (default) or TRUE. Logical indicating whether numericdist values are converted from cost-distance into Euclidean distanceusing a log-log linear regression. See alsoconvert_cdfunction.

Logical (default = TRUE) indicating whether metric valuesare returned in R (TRUE) or only stored in the patch attribute layer (FALSE)

(optional) A character string indicating the path to thedirectory that contains the project directory. It should be used when theproject directory is not in the current working directory. Default is NULL.When 'proj_path = NULL', the project directory is equal togetwd().

(optional, default = NULL) Integer or numeric valueindicating RAM gigabytes allocated to the java process. Increasing thisvalue can speed up the computations. Too large values may not be compatiblewith your machine settings.

A character string indicating the Graphab project name.The project name is also the name of the project directory in which thefile proj_name.xml is.

A character string indicating the name of the graph on whichthe modularity index is computed. This graph has been created with Graphabor usinggraphab_graph function and is associatedwith a link set. Only the links present in the graph and their correspondingweights will be used in the computation, together with patch areas.

A numeric or integer value specifying the distance at whichdispersal probability is equal toprob. This argument is mandatoryfor weighted metrics (PC, F, IF, BC, dPC, CCe, CF) but not used for others.It is used to set\alpha for computing dispersal probabilities associatedwith all inter-patch distances such that dispersal probability betweenpatches i and j isp_{ij}= e^{-\alpha d_{ij}}.

A numeric or integer value specifying the dispersal probabilityat distancedist. By default,code=0.05. It is used to set\alpha (see paramdist above).

A numeric or integer value between 0 and 1 specifying theexponent associated with patch areas in the computation of metricsweighted by patch area. By default,beta=1. Whenbeta=0, patchareas do not have any influence in the computation.

(optional, default=NULL) An integer or numeric value indicatingthe number of modules to be created. By default, it is the number thatmaximises the modularity index.

Logical (default=TRUE) indicating whether results are returnedto user.

(optional) A character string indicating the path to thedirectory that contains the project directory. It should be used when theproject directory is not in the current working directory. Default is NULL.When 'proj_path = NULL', the project directory is equal togetwd().

(optional, default = NULL) Integer or numeric valueindicating RAM gigabytes allocated to the java process. Increasing thisvalue can speed up the computations. Too large values may not be compatiblewith your machine settings.

A character string indicating the Graphab project name.The project name is also the name of the project directory in which thefile proj_name.xml is.

A character string indicating the name of the link set used.The link set is here used to get the defined cost values and compute thedistance from the point to the patches. Link sets can be createdwithgraphab_link.

Can be either;

• A character string indicating the path (absolute or relative) to ashapefile point layer

• A character string indicating the path to a .csv file with threecolumns: ID, x and y, respectively indicating the point ID, longitudeand latitude.

• A data.frame with three columns:ID, x and y, respectively indicating the point ID, longitude and latitude.

• A SpatialPointsDataFrame

The point ID column must be 'ID' by default but can also be specifiedby theid argument in all three cases.

A character string indicating the name of the column in eitherthe .csv table, data.frame or attribute table, corresponding to the IDof the points. By default, it should be 'ID'. This column is used for namingthe points when returning the output.

Logical (default=TRUE) indicating whether the metricsassociated with closest habitat patches from the points are returned tousers.

(optional) A character string indicating the path to thedirectory that contains the project directory. It should be used when theproject directory is not in the current working directory. Default is NULL.When 'proj_path = NULL', the project directory is equal togetwd().

(optional, default = NULL) Integer or numeric valueindicating RAM gigabytes allocated to the java process. Increasing thisvalue can speed up the computations. Too large values may not be compatiblewith your machine settings.

A character string indicating the Graphab project name.The project name is also the name of the project directory in which thefile proj_name.xml will be created.

A character string indicating the name of the .tif raster fileor of its path. If the path is not specified, the raster must be present inthe current working directory. Raster cell values must be in INT2S encoding.

An integer or numeric value or vector indicating thecode.s (cell value.s) of the habitat cells in the raster file.

(optional, default=FALSE) A logical indicating whethercontiguous patches corresponding to different pixel codes are merged(FALSE, default) or not merged (TRUE).Be careful, thenomerge = TRUE option is in development and we cannotguarantee the results are correct.

(optional, default=0) An integer or numeric value specifiyingthe minimum area in hectares for a habitat patch size to become a graph node.

(optional, default=NULL) An integer or numeric valuespecifying the code in the raster file associated with nodata value(often corresponding to peripheric cells)

(optional, default=NULL) An integer or numeric valuespecifying the maximum side length of the rectangular full extent of eachhabitat patch in metric units. If this side length exceedsmaxsize m,then several patches are created.(often corresponding to peripheric cells)

(optional, default=FALSE) A logical indicating whether aneighborhood of 8 pixels (TRUE) is used for patch definition. By default,con8=4, corresponding to 4 pixel neighborhood.

(optional, default = NULL) Integer or numeric valueindicating RAM gigabytes allocated to the java process. Increasing thisvalue can speed up the computations. Too large values may not be compatiblewith your machine settings.

(optional) A character string indicating the path to thedirectory that contains the project directory. It should be used when theproject directory is not in the current working directory. Default is NULL.When 'proj_path = NULL', the project directory is equal togetwd().

A character string indicating the Graphab project name.The project name is also the name of the project directory in which thefile proj_name.xml is.

A character string indicating the objects of the project thatare described. It must be either:

• mode='patches'(default): The habitat patches are describedwith synthetic descriptors (code, number, mean capacity, median capacity,capacity harmonic mean, capacity Gini coefficient) and a histogram ofcapacity distribution.

• mode='linkset': The links of a link set are describedwith synthetic descriptors (codes, costs, number, mean cost distance,median cost distance, cost distance harmonic mean, cost distance Ginicoefficient) and a histogram of cost distance distribution.

• mode='both': Both the patches and links of a linkset aredescribed

A character string indicating the name of the link setwhose properties are imported. The link set has been created with Graphabor usinggraphab_link function.

(optional) A character string indicating the path to thedirectory that contains the project directory. It should be used when theproject directory is not in the current working directory. Default is NULL.When 'proj_path = NULL', the project directory is equal togetwd().

Logical (default = FALSE) indicating whether to plot a figure ofthe resulting spatial graph. The figure is plotted using functionplot_graph_lg. The plotting can be long if the graph has manynodes and links.

Logical (default = TRUE) indicating whether the projectfeatures are returned as a list (TRUE) or only displayed in theR console (FALSE).

A character string indicating the project name. It is alsothe name of the directory in which proj_name.xml file is found. By default,'proj_name' is searched into the current working directory

A character string indicating the name of the linkset used tocreate the graph links. The linkset must have been created previously (seethe functiongraphab_link). It can be complete or planar. Thegraph is given the topology of the selected link set.

A character string indicating whether the nodes of the createdgraph are given all the attributes or metrics computed in Graphab or onlythose specific to a given graph previously created withgraphab_graphIt can be:

• nodes = "patches"(default): all the attributes and metrics ofthe habitat patches are included as node attributes inigraph object.

• nodes = "graph_name"(default): only the metrics ofthe habitat patches computed from the graph 'graph_name' created withgraphab_graph are included as node attributes inigraph object, along with some basic patch attributes.

A character string ("euclid" or "cost") indicatingwhether to weight the links with Euclidean distance orcost-distance (default) values.

(optional) A character string indicating the path to thedirectory that contains the project directory ('proj_name'). By default,'proj_name' is searched into the current working directory

Logical (default = FALSE) indicating whether to plot a figure ofthe resulting spatial graph. The figure is plotted using functionplot_graph_lg. The plotting can be long if the graph has manynodes and links.

Logical (default = FALSE) indicating whether to create an objectof classdata.frame with the node centroid spatial coordinates. Such adata.frame has 3 columns: 'ID', 'x', 'y'.

An object of classdata.frame with loci columns informatlocus (defined in packagegstudio) with as many rows asindividuals and as many columns in formatlocus as there are loci andadditional columns

A character string indicating the name of the column withpopulations' names inx

(optional) A character string indicating the name of thecolumn with individuals' ID inx

A numeric vector

A numeric value indicating the dispersal probability at a distanceequal to 'd_disp' under a negative exponential distribution.

A numeric value indicating the distance to which dispersalprobability is equal to 'p' under a negative exponential distribution.

A character string indicating the value to return:

• If 'mode = 'A” (default), the returned value 'alpha' is such thatexp(-alpha * d_disp) = p

• If 'mode = 'B”, the returned value 'alpha' is such that10(-alpha * d_disp) = p

A character string indicating the Graphab project name.The project name is also the name of the project directory in which thefile proj_name.xml is. It can be created withgraphab_project

A character string indicating the name of the first link setinvolved in the comparison. The link set has to be present in the projectand can be created withgraphab_link.

A character string indicating the name of the second link setinvolved in the comparison. The link set has to be present in the projectand can be created withgraphab_link.

(default=200) An integer or numeric valueindicating the width of the buffer created in each side of the links priorto spatial intersection. It is expressed in meters.

(optional, default=NULL) An integer or numeric valueindicating the minimum length in meters of the links to be compared. Linkswhose length is larger thanmin_length will be ignored inthe comparison.

(optional, default=NULL) A character string indicating thepath to the directory that contains the project directory. It should be usedwhen the project directory is not in the current working directory.Default is NULL. When 'proj_path = NULL', the project directory is equaltogetwd().

An object of classloci to convert

An integer indicating the ploidy level(by default, 'ploidy = 2')

A character vector indicating the coding of the allelesto be treated as missing data (by default, 'na.alleles = c("NA")')

A parameter indicating the raster file on which cost distancesare computed. It can be:

• A character string indicating the path to a raster file in format.tif or .asc.

• ARasterLayer object already loaded in R environment

All the raster cell values must be present in the column 'code' fromcost argument.

A parameter indicating the points between which cost distancesare computed. It can be either:

• A character string indicating the path to a .csv file. It must havethree columns:

  • ID: The ID of the points.

  • x: A numeric or integer indicating the longitude of the points.

  • y: A numeric or integer indicating the latitude of the points.

• Adata.frame with the spatial coordinates of the points.It must have three columns:

  • ID: The ID of the points.

  • x: A numeric or integer indicating the longitude of the points.

  • y: A numeric or integer indicating the latitude of the points.

• ASpatialPointsDataFrame with at least an attribute columnnamed "ID" with the point IDs.

The point coordinates must be in the same spatial coordinate reference systemas the raster file.

Adata.frame indicating the cost values associated to eachraster value. It must have two columns:

• 'code': raster cell values

• 'cost': corresponding cost values

A character string indicating the method used to compute thecost distances. It must be:

• 'gdistance': uses the functions from the packagegdistanceassuming that movement is possible in 8 directions from each cell, thata geo-correction is applied to correct for diagonal movement lengths and thatraster cell values correspond to resistance (and not conductance).

• 'java': uses a .jar file which is downloaded on the user's machine ifnecessary and if java is installed. This option substantially reducescomputation times and makes possible the parallelisation.

A character string indicating whether the returned object is adata.frame (return="df") or a pairwisematrix (return="mat").

An integer (4, 8, 16) indicating the directions in whichmovement can take place from a cell. Only used whenmethod="gdistance".By default,direction=8.

An integer indicating how many computer cores are usedto run the .jar file. By default,parallel.java=1.

(optional, default = NULL) Integer or numeric valueindicating RAM gigabytes allocated to the java process when used. Increasingthis value can speed up the computations. Too large values may not becompatible with your machine settings.

An object of classgenind that contains the multilocusgenotypes (format 'locus') of the individuals as well as their populations.

A character string indicating the method used to compute themultilocus genetic distance between populations

• If 'dist = 'basic” (default), then the multilocus genetic distance iscomputed using a formula of Euclidean geneticdistance (Excoffier et al., 1992)

• If 'dist = 'weight”, then the multilocus genetic distance is computedas in Fortuna et al. (2009). It is a Euclidean genetic distance giving moreweight to rare alleles

• If 'dist = 'PG”, then the multilocus genetic distance is computed asin popgraph::popgraph function, following several steps of PCA and SVD(Dyer et Nason, 2004).

• If 'dist = 'DPS”, then the genetic distance used is equal to1 - the proportion of shared alleles (Bowcock, 1994)

• If 'dist = 'FST”, then the genetic distance used is the pairwiseFST (Weir et Cockerham, 1984)

• If 'dist = 'FST_lin”, then the genetic distance used is the linearisedpairwise FST (Weir et Cockerham, 1984)(FST_lin = FST/(1-FST))

• If 'dist = 'PCA”, then the genetic distance is computed following aPCA of the matrix of allelic frequencies by population. It is aEuclidean genetic distance between populations in the multidimensionalspace defined by all the independent principal components.

• If 'dist = 'GST”, then the genetic distance used is theG'ST (Hedrick, 2005). See graph4lg <= 1.6.0 only, because it used diveRsity

• If 'dist = 'D”, then the genetic distance used isJost's D (Jost, 2008). See graph4lg <= 1.6.0 only, because it used diveRsity

(optional) Logical. Should negative and null FST, FST_lin,GST or D values be replaced by half the minimum positive value?This option allows to compute Gabriel graphs from these "distances".Default is null_val = FALSE.This option only works if 'dist = 'FST” or 'FST_lin' or 'GST' or 'D'

An object of class :

• data.frame with 3 columns: 2 columns with the point spatialcoordinates and another column with point IDs

• SpatialPointsDataFrame

(ifdata is of classdata.frame) A character stringindicating the name of the column ofdata with the point IDs

(ifdata is of classdata.frame) A character stringindicating the name of the column ofdata with the point longitude

(ifdata is of classdata.frame) A character stringindicating the name of the column ofdata with the point latitude

A character string indicating the type of coordinatereference system:

• 'proj' (default): a projected coordinate reference system

• 'polar': a polar coordinate reference system, such as WGS84

A character string indicating the formula used to computethe Great Circle distance:

• 'vicenty'(default): Vincenty inverse formula for ellipsoids

• 'slc': Spherical Law of Cosines

• 'hvs': Harversine formula

An object of classgenind

An object of classgenind

A RasterLayer object corresponding to a categorical raster layer

An integer value or vector with the value(s) corresponding tothe code values of the raster layer within which points will be sampled.are computed.

An integer value indicating the width of the edge(in meters) of the raster layer which is ignored during the sampling(default = 0). It prevents from sampling in the margins of the study area.

An integer value indicating which cells are consideredadjacent when contiguous patches are delineated (it should be 8(default, Queen's case) or 4 (Rook's case)). This parameter is ignoredwhenby_patch = FALSE.

An integer value indicating the minimum surface of a patchconsidered for the sampling in number of raster cells. This parameter is usedwhatever theby_patch argument is. Default is 0.

A graph object of classigraph

(optional, default = NULL) If 'mode = 'spatial”, it is adata.frame with the spatial coordinates of the graph nodes.It must have three columns :

• ID: A character string indicating the name of the graph nodes.The names must be the same as the node names of the graph ofclassigraph (igraph::V(graph)$name)

• x: A numeric or integer indicating the longitude of the graph nodes.

• y: A numeric or integer indicating the latitude of the graph nodes.

This argument is not used when 'mode = 'aspatial” and mandatory when 'mode ='spatial”.

A character string indicating whether the graph isspatial ('mode = 'spatial”) or not ('mode = 'aspatial” (default))

(optional, default = NULL) A character string indicatingwhether the links of the graph are weighted by distances or by similarityindices. It is only used when 'mode = 'aspatial” to compute the nodepositions with Fruchterman and Reingold algorithm. It can be equal to:

• 'distance': Link weights correspond to distances. Nodes that are closeto each other will be close on the figure.

• 'similarity': Link weights correspond to similarity indices. Nodes thatare similar to each other will be close on the figure.

(optional, default = NULL) A character string indicatinghow the width of the link is set on the figure. Their width can be:

• inversely proportional to link weights ("inv_w", convenient withdistances, default)

• proportional to link weights ("w")

(optional, default = NULL) A character string indicatingthe graph node attribute used to set the node size on the figure. It must bethe name of a numeric or integer node attribute from the graph.

(optional, default = NULL) A character string indicatingthe graph node modules used to set the node color on the figure. It must bethe name of a node attribute from the graph with discrete values.

(optional, default = NULL) A character string indicating thecolor used to plot the nodes (default: "#F2B950"). It must be a hexadecimalcolor code or a color used by default in R. It cannot be used if 'module' isspecified.

A graph object of classigraph whose links are weighted

A character string indicating the color used to fillthe bars (default: "#396D35"). It must be a hexadecimal color code ora color used by default in R.

(default values: NULL) A numeric or an integer specifyingthe width of the classes displayed on the histogram. When it is notspecified, the width is equal to the difference between the minimum andmaximum values divided by 80.

An object of classgenindfrom packageadegenet.

(optional) A character vector indicating population names.It is of the same length as the number of populations. Without thisargument, populations are given the names they have initially in the'genind' object (which is sometimes only a number). The order of thepopulation names must match with their order in the 'genind' object.The function does not reorder them. Users must be careful.

(optional) A character vector indicating the population-levelindices to compute. These indices can be:

• Mean allelic richness by locus bypopulation (indices = c("A", ...))

• Mean expected heterozygosity by locus bypopulation (indices = c("He",...))

• Mean observed heterozygosity by locus bypopulation (indices = c("Ho",...))

• Number of individuals bypopulation (indices = c("Nb_ind", ...))

• Total allelic richness bypopulation (indices = c("A_tot",...))

By default,indices = c("Nb_ind", "A", "He", "Ho").

An object of classgenindfrom packageadegenet.

(optional default = NULL) The maximum standardized sample sizeused by ADZE software (MAX_G) in ADZE manual. It is equal to twice theminimum number of individuals considered for the rarefaction analysis. Bydefault, it is equal to twice the number of individuals in the smallestpopulation. Ohterwise, it must be either a numeric or integer value.

(optional) A character vector indicating population names.It is of the same length as the number of populations. Without thisargument, populations are given the names they have initially in the'genind' object (which is sometimes only a number). The order of thepopulation names must match with their order in the 'genind' object.The function does not reorder them. Users must be careful.

A character string indicating whether you use a Linux ('linux')or Windows ('win') operating system.

A pairwise matrix which can be:

• An object of classmatrix. It must havethe same row names and column names. If values represent distances,diagonal elements should be equal to 0.

• An object of classdist. In that, its column numbers areused to create IDs in the resulting data.frame.

An object of classmatrix

A character vector with the rows and columns names of the matrixin the order in which they will be ordered by the function. All its elementsmust be rows and columns names of the matrixmat.

A RasterLayer object corresponding to a categorical raster layer

An integer value or vector with the value(s) corresponding tothe code values of the raster layer within which points will be sampled.

An integer value indicating the number of points to be sampled

An integer value indicating the minimum distance separatingthe sampled points (default = 0).

An integer value indicating the width of the edge of theraster layer which is ignored during the sampling (default = 0). It preventsfrom sampling in the margins of the study area.

A logical value indicating whether contiguous patches withcells having the same code value are delineated prior tosampling (default = TRUE). It prevents from sampling several points in thesame contiguous patch.

An integer value indicating which cells are consideredadjacent when contiguous patches are delineated (it should be 8(default, Queen's case) or 4 (Rook's case)). This parameter is ignoredwhenby_patch = FALSE.

An integer value indicating the minimum surface of a patchconsidered for the sampling in number of raster cells. This parameter is usedwhatever theby_patch argument is. Default is 0.

A logical value indicating whether sampling in large patchesis more likely (default = TRUE). Ifby_patch = FALSE, this parameteris ignored. Whenprop_area = TRUE, the probability to sample a givenpatch is proportional to its area.

An integer value indicating how many sampling steps areperformed to identify a point set satisfying all the conditions beforereturning an error.

A character string indicating the type of returned output:

• 'data.frame': Adata.frame with three/four columns:

  • ID: The point or patch centroid ID

  • x: The point or patch centroid longitude

  • y: The point or patch centroid latitude

  • area: The area of the sampled patch (only ifby_patch = TRUE)

• 'pts_layer': ASpatialPointsDataFrame layer correspondingto the sampled point (points or patch centroids)

• 'poly_layer': ASpatialPolygonsDataFrame layer correspondingto the sampled patch polygons

A logical value indicating whether the result should bedescribed or not (default = FALSE). Ifdesc = TRUE, then the Ginicoefficient of the distances between points and of the patch areas (ifby_patch = TRUE) is computed with thegini_coeff.An histogram of the link weights is also described.

Numeric or integer vector

A symmetricmatrix ordist object with pairwisegenetic distances between populations or sample sites.

A symmetricmatrix ordist object with pairwiselandscape distances between populations or sample sites. These distancescan be Euclidean distances, cost-distances or resistance distances,among others.

A character string indicating the smoothing methodused to fit a line on the scatterplot. Possible values are the same aswith function 'geom_smooth()' fromggplot2 : 'lm', 'glm', 'gam','loess' (default).

(optional) A numeric or integer value used to remove valuesfrom the data before to plot. All genetic distances values abovethr_gd are removed from the data.

(optional) A numeric or integer value used to remove valuesfrom the data before to plot. All landscape distances values abovethr_ld are removed from the data.

Logical (optional, default = TRUE) indicating whether theconfidence interval around the smooth line is displayed.

(optional) A character string indicating the colorused to plot the smoothing line (default: "blue"). It must be a hexadecimalcolor code or a color used by default in R.

(optional) Character string indicating the colorused to plot the points (default: "#999999"). It must be a hexadecimal colorcode or a color used by default in R.

A symmetric (complete)matrix ordist object withpairwise (genetic or landscape) distances between populations or samplesites. These values will be the point coordinates on the y axis.mat_y is the matrix used to weight the links of the graphx,whose nodes correspond to row and column names ofmat_y.

A symmetric (complete)matrix ordist object withpairwise (genetic or landscape) distances between populations or samplesites. These values will be the point coordinates on the x axis.mat_x andmat_y must have the same row and column names,ordered in the same way.

A graph object of classigraph.Its nodes must have the same names as the row and column ofmat_y andmat_x matrices.x must have weighted links.Link weights have to be values frommat_y matrix.graph mustbe an undirected graph.

(optional) A numeric or integer value used to remove valuesfrom the data before to plot. All values frommat_y abovethr_yare removed from the data.

(optional) A numeric or integer value used to remove valuesfrom the data before to plot. All values frommat_x abovethr_xare removed from the data.

(optional) A character string indicating the color used toplot the points associated to all populations or sample sitespairs (default: "#999999"). It must be a hexadecimal colorcode or a color used by default in R.

(optional) A character string indicating the color used toplot the points associated to populations or sample sites pairs connected onthe graph (default: "black"). It must be a hexadecimal colorcode or a color used by default in R.

A character string indicating the path to the STRUCTURE file informat .txt, or alternatively the name of the file in the working directory.The STRUCTURE file must only have :

• A first column with the IDs of the individuals(can be a simple number)

• A second column with the IDs of the populations(can be a simple number)

• Some loci columns : as many columns as loci in the data

The row for loci names is optional but recommended.Each individual is displayed on 2 rows.

(optional) A character vector indicating the populationnames in the same order as in the STRUCTURE file. It is of the same lengthas the number of populations. Without this argument, populations arenumbered from 1 to the total number of individuals.

A character vector with the names of the loci if notspecified in the file first row. This argument is mandatory if theSTRUCTURE file does not include the names of the loci in the first row.In other cases, the names of the loci is extracted from the file first row

(optional) A character vector indicating the individualnames in the same order as in the STRUCTURE file. It is of the same lengthas the number of individuals. Without this argument, individuals arenumbered from 1 to the total number of individuals.