[igraph] Node Overlap and Positioning

[Lorenzo Isella]

Dear All,
I have not used the igraph in a long time, so I am a bit rusty.
Essentially, I would like to make a nice eye-catching plot of a simple
network.
Please have a look at

https://ged-project.de/research/studies/factory-europe/

where if you scroll down a bit you find a nice visualization.
I like it because

1) there is no overlapping between the nodes
2) the bigger nodes have a rather central position in the network plot
3) the edges are curved in a very smooth way.


As to 1), from my googling, it appears that igraph has no concept of node
size, so you have to fine tune the layout, xlim,ylim and node size
yourself. Or has anything changed at this regard?

For 2), what is the recommended layout?

For 3), yes, I can bend the edges, but the result is not nearly as smooth
as in that network visualization.
To fix the ideas, please have a look at the self contained example at the
end of the email.
Any suggestion is appreciated.
Cheers

Lorenzo


##########################################################

library(igraph)

g <- structure(list(28, FALSE, c(6, 6, 8, 9, 10, 10, 11, 12, 13, 14,
15, 16, 16, 17, 17, 17, 17, 17, 17, 18, 18, 19, 19, 19, 20, 20,
20, 20, 20, 20, 21, 21, 21, 22, 22, 22, 23, 23, 23, 24, 24, 24,
25, 25, 25, 25, 25, 25, 26, 26, 26, 27, 27, 27), c(1, 2, 2, 3,
2, 7, 6, 1, 4, 10, 2, 1, 12, 1, 3, 5, 6, 10, 13, 7, 15, 14, 17,
18, 0, 1, 2, 14, 18, 19, 0, 6, 7, 0, 9, 18, 17, 20, 22, 5, 7,
11, 0, 8, 10, 18, 22, 24, 6, 9, 14, 4, 21, 25), c(0, 1, 2, 3,
4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20,
21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36,
37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52,
53), c(24, 30, 33, 42, 0, 7, 11, 13, 25, 1, 2, 4, 10, 26, 3,
14, 8, 51, 15, 39, 6, 16, 31, 48, 5, 19, 32, 40, 43, 34, 49,
9, 17, 44, 41, 12, 18, 21, 27, 50, 20, 22, 36, 23, 28, 35, 45,
29, 37, 52, 38, 46, 47, 53), c(0, 0, 0, 0, 0, 0, 0, 2, 2, 3,
4, 6, 7, 8, 9, 10, 11, 13, 19, 21, 24, 30, 33, 36, 39, 42, 48,
51, 54), c(0, 4, 9, 14, 16, 18, 20, 24, 28, 29, 31, 34, 35, 36,
37, 40, 41, 41, 43, 47, 48, 49, 50, 52, 52, 53, 54, 54, 54),
list(c(1, 0, 1), structure(list(name = "Erdos renyi (gnp) graph",
type = "gnp", loops = FALSE, p = 0.142857142857143), .Names =
c("name",
"type", "loops", "p")), structure(list(names = c("BE", "BG",
"CZ", "DK", "DE", "EE", "IE", "EL", "ES", "FR", "IT", "CY",
"LV", "LT", "LU", "HU", "MT", "NL", "AT", "PL", "PT", "RO",
"SI", "SK", "FI", "SE", "UK", "HR")), .Names = "names"),
structure(list(weight = c(16.3104184714332, 30.7293642773093,
1.51651715633125, 46.5498389394643, 117.400128622748,
9116.96298628885, 1101.62292417749, 12.3449820775232,
158.642126867059, 1.81516408646891, 760.982895191571,
1011.05564809154, 31677.1638228568, 189.344972592696,
13.1424122833109, 66.5517926330624, 12.0922729908787,
4.58139821481045, 90.1042955878783, 3.71064650396136,
90.4281997569113, 158.365362288124, 46.6972207187443,
30.3317841283049, 7.78960961626955, 1.95560245248322,
697.72330148994, 58.3866681241721, 660.780592399842,
1.30621866244087, 90.6566452600129, 411.372413931909,
50.4594917412583, 30.747955195432, 5.22931377396076,
26225.4597289607, 518.792688616195, 12998.9635603034,
69.4202716300456, 8.21344218621299, 0.582827301648975,
8.09867531337074, 107.652085596974, 1142.61073604216,
116.541846414264, 1.62280326041373, 405.914678311703,
0.386623934247936, 18.218603738158, 21.1501080346551,
0.158066706508139, 862.791072359265, 8.42653129127126,
20.0432021225717)), .Names = "weight"))), class = "igraph")

g <- upgrade_graph(g)

## l <- layout.fruchterman.reingold(g)

l <- layout_with_dh(g)

l <- layout.norm(l, -1,1, -1,1)


l <- norm_coords(layout.graphopt(g), xmin=-2, xmax=2, ymin=-2, ymax=2)


set.seed(1234)
pdf("toy-network.pdf")

plot(g, layout=l,
## rescale=F, ylim=c(0.4,2.6), xlim=c(-3.4, -1.7),
## vertex.label.dist=0.5,
vertex.color=adjustcolor("blue", .5),
vertex.frame.color="#ff000033",
edge.color="#55555533",
vertex.label=V(g)$names,
vertex.label.color="black",
vertex.size=log(strength(g))*2,
edge.width= log(1+E(g)$weight)
, edge.curved=0.3,
edge.label=NA )


dev.off()

[Sebastián Goinheix]

I think the central nodes are there because of their location, not because
of their centrality.
Gephi has a layout with geographic coordinates. You can export the database
to gexf format with the igraph.to.gexf function ("rgexf" package). I have
not used it but I guess it works.


--------

*Sebastián Goinheix*Instituto de Economía - UdelaR
Tel: [phone] (ext. 2416)
*Por favor, evite imprimir este correo.*

Post by Lorenzo Isella
Dear All,
I have not used the igraph in a long time, so I am a bit rusty.
Essentially, I would like to make a nice eye-catching plot of a simple
network.
Please have a look at
https://ged-project.de/research/studies/factory-europe/
where if you scroll down a bit you find a nice visualization.
I like it because
1) there is no overlapping between the nodes
2) the bigger nodes have a rather central position in the network plot
3) the edges are curved in a very smooth way.
As to 1), from my googling, it appears that igraph has no concept of node
size, so you have to fine tune the layout, xlim,ylim and node size
yourself. Or has anything changed at this regard?
For 2), what is the recommended layout?
For 3), yes, I can bend the edges, but the result is not nearly as smooth
as in that network visualization.
To fix the ideas, please have a look at the self contained example at the
end of the email.
Any suggestion is appreciated.
Cheers
Lorenzo
##########################################################
library(igraph)
g <- structure(list(28, FALSE, c(6, 6, 8, 9, 10, 10, 11, 12, 13, 14,
15, 16, 16, 17, 17, 17, 17, 17, 17, 18, 18, 19, 19, 19, 20, 20,
20, 20, 20, 20, 21, 21, 21, 22, 22, 22, 23, 23, 23, 24, 24, 24,
25, 25, 25, 25, 25, 25, 26, 26, 26, 27, 27, 27), c(1, 2, 2, 3,
2, 7, 6, 1, 4, 10, 2, 1, 12, 1, 3, 5, 6, 10, 13, 7, 15, 14, 17,
18, 0, 1, 2, 14, 18, 19, 0, 6, 7, 0, 9, 18, 17, 20, 22, 5, 7,
11, 0, 8, 10, 18, 22, 24, 6, 9, 14, 4, 21, 25), c(0, 1, 2, 3,
4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20,
21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36,
37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52,
53), c(24, 30, 33, 42, 0, 7, 11, 13, 25, 1, 2, 4, 10, 26, 3,
14, 8, 51, 15, 39, 6, 16, 31, 48, 5, 19, 32, 40, 43, 34, 49,
9, 17, 44, 41, 12, 18, 21, 27, 50, 20, 22, 36, 23, 28, 35, 45,
29, 37, 52, 38, 46, 47, 53), c(0, 0, 0, 0, 0, 0, 0, 2, 2, 3,
4, 6, 7, 8, 9, 10, 11, 13, 19, 21, 24, 30, 33, 36, 39, 42, 48,
51, 54), c(0, 4, 9, 14, 16, 18, 20, 24, 28, 29, 31, 34, 35, 36,
37, 40, 41, 41, 43, 47, 48, 49, 50, 52, 52, 53, 54, 54, 54),
list(c(1, 0, 1), structure(list(name = "Erdos renyi (gnp) graph",
type = "gnp", loops = FALSE, p = 0.142857142857143), .Names =
c("name",
"type", "loops", "p")), structure(list(names = c("BE", "BG",
"CZ", "DK", "DE", "EE", "IE", "EL", "ES", "FR", "IT", "CY",
"LV", "LT", "LU", "HU", "MT", "NL", "AT", "PL", "PT", "RO",
"SI", "SK", "FI", "SE", "UK", "HR")), .Names = "names"),
structure(list(weight = c(16.3104184714332, 30.7293642773093,
1.51651715633125, 46.5498389394643, 117.400128622748,
9116.96298628885, 1101.62292417749, 12.3449820775232,
158.642126867059, 1.81516408646891, 760.982895191571,
1011.05564809154, 31677.1638228568, 189.344972592696,
13.1424122833109, 66.5517926330624, 12.0922729908787,
4.58139821481045, 90.1042955878783, 3.71064650396136,
90.4281997569113, 158.365362288124, 46.6972207187443,
30.3317841283049, 7.78960961626955, 1.95560245248322,
697.72330148994, 58.3866681241721, 660.780592399842,
1.30621866244087, 90.6566452600129, 411.372413931909,
50.4594917412583, 30.747955195432, 5.22931377396076,
26225.4597289607, 518.792688616195, 12998.9635603034,
69.4202716300456, 8.21344218621299, 0.582827301648975,
8.09867531337074, 107.652085596974, 1142.61073604216,
116.541846414264, 1.62280326041373, 405.914678311703,
0.386623934247936, 18.218603738158, 21.1501080346551,
0.158066706508139, 862.791072359265, 8.42653129127126,
20.0432021225717)), .Names = "weight"))), class = "igraph")
g <- upgrade_graph(g)
## l <- layout.fruchterman.reingold(g)
l <- layout_with_dh(g)
l <- layout.norm(l, -1,1, -1,1)
l <- norm_coords(layout.graphopt(g), xmin=-2, xmax=2, ymin=-2, ymax=2)
set.seed(1234)
pdf("toy-network.pdf")
plot(g, layout=l,
## rescale=F, ylim=c(0.4,2.6), xlim=c(-3.4, -1.7),
## vertex.label.dist=0.5,
vertex.color=adjustcolor("blue", .5),
vertex.frame.color="#ff000033",
edge.color="#55555533",
vertex.label=V(g)$names,
vertex.label.color="black",
vertex.size=log(strength(g))*2,
edge.width= log(1+E(g)$weight)
, edge.curved=0.3,
edge.label=NA )
dev.off()
_______________________________________________
igraph-help mailing list
https://lists.nongnu.org/mailman/listinfo/igraph-help

[Esteban Moro]

Hi Lorenzo.

As far as I can see, the layout in the visualization corresponds to the geographical coordinates for each country. So that’s why nodes have no overlap and bigger nodes (France, Italy, GBR, DEU) appear in the center

Best


Esteban Moro Egido
Visiting Professor, MIT Media Lab
Associate Professor, Universidad Carlos III de Madrid
estebanmoro.org <http://estebanmoro.org/>

Post by Lorenzo Isella
Dear All,
I have not used the igraph in a long time, so I am a bit rusty.
Essentially, I would like to make a nice eye-catching plot of a simple network.
Please have a look at
https://ged-project.de/research/studies/factory-europe/ <https://ged-project.de/research/studies/factory-europe/>
where if you scroll down a bit you find a nice visualization.
I like it because
1) there is no overlapping between the nodes
2) the bigger nodes have a rather central position in the network plot
3) the edges are curved in a very smooth way.
As to 1), from my googling, it appears that igraph has no concept of node size, so you have to fine tune the layout, xlim,ylim and node size yourself. Or has anything changed at this regard?
For 2), what is the recommended layout?
For 3), yes, I can bend the edges, but the result is not nearly as smooth as in that network visualization.
To fix the ideas, please have a look at the self contained example at the end of the email.
Any suggestion is appreciated.
Cheers
Lorenzo
##########################################################
library(igraph)
g <- structure(list(28, FALSE, c(6, 6, 8, 9, 10, 10, 11, 12, 13, 14,
15, 16, 16, 17, 17, 17, 17, 17, 17, 18, 18, 19, 19, 19, 20, 20,
20, 20, 20, 20, 21, 21, 21, 22, 22, 22, 23, 23, 23, 24, 24, 24,
25, 25, 25, 25, 25, 25, 26, 26, 26, 27, 27, 27), c(1, 2, 2, 3,
2, 7, 6, 1, 4, 10, 2, 1, 12, 1, 3, 5, 6, 10, 13, 7, 15, 14, 17,
18, 0, 1, 2, 14, 18, 19, 0, 6, 7, 0, 9, 18, 17, 20, 22, 5, 7,
11, 0, 8, 10, 18, 22, 24, 6, 9, 14, 4, 21, 25), c(0, 1, 2, 3,
4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20,
21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36,
37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52,
53), c(24, 30, 33, 42, 0, 7, 11, 13, 25, 1, 2, 4, 10, 26, 3,
14, 8, 51, 15, 39, 6, 16, 31, 48, 5, 19, 32, 40, 43, 34, 49,
9, 17, 44, 41, 12, 18, 21, 27, 50, 20, 22, 36, 23, 28, 35, 45,
29, 37, 52, 38, 46, 47, 53), c(0, 0, 0, 0, 0, 0, 0, 2, 2, 3,
4, 6, 7, 8, 9, 10, 11, 13, 19, 21, 24, 30, 33, 36, 39, 42, 48,
51, 54), c(0, 4, 9, 14, 16, 18, 20, 24, 28, 29, 31, 34, 35, 36,
37, 40, 41, 41, 43, 47, 48, 49, 50, 52, 52, 53, 54, 54, 54),
list(c(1, 0, 1), structure(list(name = "Erdos renyi (gnp) graph",
type = "gnp", loops = FALSE, p = 0.142857142857143), .Names = c("name",
"type", "loops", "p")), structure(list(names = c("BE", "BG",
"CZ", "DK", "DE", "EE", "IE", "EL", "ES", "FR", "IT", "CY",
"LV", "LT", "LU", "HU", "MT", "NL", "AT", "PL", "PT", "RO",
"SI", "SK", "FI", "SE", "UK", "HR")), .Names = "names"),
structure(list(weight = c(16.3104184714332, 30.7293642773093,
1.51651715633125, 46.5498389394643, 117.400128622748,
9116.96298628885, 1101.62292417749, 12.3449820775232,
158.642126867059, 1.81516408646891, 760.982895191571,
1011.05564809154, 31677.1638228568, 189.344972592696,
13.1424122833109, 66.5517926330624, 12.0922729908787,
4.58139821481045, 90.1042955878783, 3.71064650396136,
90.4281997569113, 158.365362288124, 46.6972207187443,
30.3317841283049, 7.78960961626955, 1.95560245248322,
697.72330148994, 58.3866681241721, 660.780592399842,
1.30621866244087, 90.6566452600129, 411.372413931909,
50.4594917412583, 30.747955195432, 5.22931377396076,
26225.4597289607, 518.792688616195, 12998.9635603034,
69.4202716300456, 8.21344218621299, 0.582827301648975,
8.09867531337074, 107.652085596974, 1142.61073604216,
116.541846414264, 1.62280326041373, 405.914678311703,
0.386623934247936, 18.218603738158, 21.1501080346551,
0.158066706508139, 862.791072359265, 8.42653129127126,
20.0432021225717)), .Names = "weight"))), class = "igraph")
g <- upgrade_graph(g)
## l <- layout.fruchterman.reingold(g)
l <- layout_with_dh(g)
l <- layout.norm(l, -1,1, -1,1)
l <- norm_coords(layout.graphopt(g), xmin=-2, xmax=2, ymin=-2, ymax=2)
set.seed(1234)
pdf("toy-network.pdf")
plot(g, layout=l,
## rescale=F, ylim=c(0.4,2.6), xlim=c(-3.4, -1.7),
## vertex.label.dist=0.5,
vertex.color=adjustcolor("blue", .5),
vertex.frame.color="#ff000033",
edge.color="#55555533",
vertex.label=V(g)$names,
vertex.label.color="black",
vertex.size=log(strength(g))*2,
edge.width= log(1+E(g)$weight)
, edge.curved=0.3,
edge.label=NA )
dev.off()
_______________________________________________
igraph-help mailing list
https://lists.nongnu.org/mailman/listinfo/igraph-help

[George Vega Yon]

This post might be inappropriate. Click to display it.

[Lorenzo Isella]

Thanks everybody for their help!
I will try out your suggestions and post again if I have trouble.
Just for the record, being an avid ggplot2 user, I am also considering
ggraph for making panel plots of networks, but this is another story.
Cheers

Lorenzo

Post by George Vega Yon
Lorenzo,
I suggest you take a look at the rgexf package (which I maintain), in
particular, the development version https://github.com/gvegayon/rgexf
which has a bunch of new features including an integration with the gexf-js
javascript library which allows you to get Gephi-like viz on the browser.
That said, here is an example of a couple of things that I would do with
1. Create a new plotting device, and on top of that add the igraph plot.
That gives you more control and allows you to, say, not restrict yourself
to the square layout.
2. Use either white or some other color for the frame
3. Tryout netdiffuseR::igraph_vertex_rescale function. This function from
the netdiffuseR R package, which I maintain as well, allows rescaling
vertex size so that instead of using absolute values you can define how
much of the plotting device, in percentage, will the largest and smallest
node occupy.
Below an example including the plot
HIH
library(igraph)
g <- structure(list(28, FALSE, c(6, 6, 8, 9, 10, 10, 11, 12, 13, 14,
15, 16, 16, 17, 17, 17, 17, 17, 17, 18,
18, 19, 19, 19, 20, 20,
20, 20, 20, 20, 21, 21, 21, 22, 22, 22,
23, 23, 23, 24, 24, 24,
25, 25, 25, 25, 25, 25, 26, 26, 26, 27,
27, 27), c(1, 2, 2, 3,
2, 7, 6, 1, 4, 10, 2, 1, 12, 1, 3, 5, 6, 10, 13, 7, 15, 14, 17,
18, 0, 1, 2, 14, 18, 19, 0, 6, 7, 0, 9, 18, 17, 20, 22, 5, 7,
11, 0, 8, 10, 18, 22, 24, 6, 9, 14, 4, 21, 25), c(0, 1, 2, 3,
4, 5, 6, 7, 8,
9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20,
21, 22, 23, 24,
25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36,
37, 38, 39, 40,
41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52,
53), c(24, 30,
33, 42, 0, 7, 11, 13, 25, 1, 2, 4, 10, 26, 3,
14, 8,
51, 15, 39, 6, 16, 31, 48, 5, 19, 32, 40, 43, 34, 49,
9, 17,
44, 41, 12, 18, 21, 27, 50, 20, 22, 36, 23, 28, 35, 45,
29, 37,
52, 38, 46, 47, 53), c(0, 0, 0, 0, 0, 0, 0, 2, 2, 3,
4, 6, 7, 8, 9, 10, 11, 13, 19, 21, 24, 30, 33, 36, 39, 42, 48,
51, 54), c(0, 4, 9, 14, 16, 18, 20, 24, 28, 29, 31, 34, 35, 36,
37, 40, 41, 41, 43, 47, 48, 49, 50, 52,
52, 53, 54, 54, 54),
list(c(1, 0, 1), structure(list(name = "Erdos renyi (gnp) graph",
type = "gnp", loops =
FALSE, p = 0.142857142857143), .Names = c("name",
"type", "loops", "p")),
structure(list(names = c("BE", "BG",
"CZ", "DK", "DE", "EE", "IE", "EL", "ES", "FR", "IT", "CY",
"LV", "LT", "LU", "HU", "MT", "NL", "AT", "PL", "PT", "RO",
"SI", "SK", "FI", "SE", "UK", "HR")), .Names = "names"),
structure(list(weight = c(16.3104184714332, 30.7293642773093,
1.51651715633125,
46.5498389394643, 117.400128622748,
9116.96298628885,
1101.62292417749, 12.3449820775232,
158.642126867059,
1.81516408646891, 760.982895191571,
1011.05564809154,
31677.1638228568, 189.344972592696,
13.1424122833109,
66.5517926330624, 12.0922729908787,
4.58139821481045,
90.1042955878783, 3.71064650396136,
90.4281997569113,
158.365362288124, 46.6972207187443,
30.3317841283049,
7.78960961626955, 1.95560245248322,
697.72330148994,
58.3866681241721, 660.780592399842,
1.30621866244087,
90.6566452600129, 411.372413931909,
50.4594917412583,
30.747955195432, 5.22931377396076,
26225.4597289607,
518.792688616195, 12998.9635603034,
69.4202716300456,
8.21344218621299, 0.582827301648975,
8.09867531337074,
107.652085596974, 1142.61073604216,
116.541846414264,
1.62280326041373, 405.914678311703,
0.386623934247936,
18.218603738158, 21.1501080346551,
0.158066706508139,
862.791072359265, 8.42653129127126,
20.0432021225717)),
.Names = "weight"))), class = "igraph")
g <- upgrade_graph(g)
## l <- layout.fruchterman.reingold(g)
l <- layout_with_dh(g)
l <- layout.norm(l, -1,1, -1,1)
l <- norm_coords(layout.graphopt(g), xmin=-2, xmax=2, ymin=-2, ymax=2)
set.seed(1234)
# pdf("toy-network.pdf")
# I rather puttin the plot on a new plotting device
oldpar <- par(no.readonly = TRUE)
par(
mar = rep(2,4),
xpd = NA # This ensures you don't clip the images
)
plot.new()
plot.window(
xlim = range(l[,1]),
ylim = range(l[,2])
)
plot(g, layout=l,
## rescale=F, ylim=c(0.4,2.6), xlim=c(-3.4, -1.7),
## vertex.label.dist=0.5,
vertex.color=adjustcolor("blue", .5),
# White works nicer for the borders
# vertex.frame.color="#ff000033",
vertex.frame.color= "white",
edge.color="#55555533",
vertex.label=V(g)$names,
vertex.label.color="black",
# You can use netdiffuseR
# vertex.size=log(strength(g))*2,
vertex.size = netdiffuseR::igraph_vertex_rescale(
vertex.size = strength(g),
minmax.relative.size = c(.025, .1)
),
edge.width= log(1+E(g)$weight)
, edge.curved=0.3,
edge.label=NA,
# NEW OPTIONS
add = TRUE,
rescale=FALSE
)
par(oldpar)
# dev.off()
George G. Vega Yon
[phone] <+1%20626-381-8171>
http://ggvy.cl

Post by Esteban Moro
Hi Lorenzo.
As far as I can see, the layout in the visualization corresponds to the
geographical coordinates for each country. So that’s why nodes have no
overlap and bigger nodes (France, Italy, GBR, DEU) appear in the center
Best
*Esteban Moro Egido*Visiting Professor, *MIT Media Lab*
Associate Professor, *Universidad Carlos III de Madrid*
estebanmoro.org
Dear All,
I have not used the igraph in a long time, so I am a bit rusty.
Essentially, I would like to make a nice eye-catching plot of a simple network.
Please have a look at
https://ged-project.de/research/studies/factory-europe/
where if you scroll down a bit you find a nice visualization.
I like it because
1) there is no overlapping between the nodes
2) the bigger nodes have a rather central position in the network plot
3) the edges are curved in a very smooth way.
As to 1), from my googling, it appears that igraph has no concept of node
size, so you have to fine tune the layout, xlim,ylim and node size
yourself. Or has anything changed at this regard?
For 2), what is the recommended layout?
For 3), yes, I can bend the edges, but the result is not nearly as smooth
as in that network visualization.
To fix the ideas, please have a look at the self contained example at the
end of the email.
Any suggestion is appreciated.
Cheers
Lorenzo
##########################################################
library(igraph)
g <- structure(list(28, FALSE, c(6, 6, 8, 9, 10, 10, 11, 12, 13, 14,
15, 16, 16, 17, 17, 17, 17, 17, 17, 18, 18, 19, 19, 19, 20, 20,
20, 20, 20, 20, 21, 21, 21, 22, 22, 22, 23, 23, 23, 24, 24, 24,
25, 25, 25, 25, 25, 25, 26, 26, 26, 27, 27, 27), c(1, 2, 2, 3,
2, 7, 6, 1, 4, 10, 2, 1, 12, 1, 3, 5, 6, 10, 13, 7, 15, 14, 17,
18, 0, 1, 2, 14, 18, 19, 0, 6, 7, 0, 9, 18, 17, 20, 22, 5, 7,
11, 0, 8, 10, 18, 22, 24, 6, 9, 14, 4, 21, 25), c(0, 1, 2, 3,
4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20,
21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36,
37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52,
53), c(24, 30, 33, 42, 0, 7, 11, 13, 25, 1, 2, 4, 10, 26, 3,
14, 8, 51, 15, 39, 6, 16, 31, 48, 5, 19, 32, 40, 43, 34, 49,
9, 17, 44, 41, 12, 18, 21, 27, 50, 20, 22, 36, 23, 28, 35, 45,
29, 37, 52, 38, 46, 47, 53), c(0, 0, 0, 0, 0, 0, 0, 2, 2, 3,
4, 6, 7, 8, 9, 10, 11, 13, 19, 21, 24, 30, 33, 36, 39, 42, 48,
51, 54), c(0, 4, 9, 14, 16, 18, 20, 24, 28, 29, 31, 34, 35, 36,
37, 40, 41, 41, 43, 47, 48, 49, 50, 52, 52, 53, 54, 54, 54),
list(c(1, 0, 1), structure(list(name = "Erdos renyi (gnp) graph",
type = "gnp", loops = FALSE, p = 0.142857142857143), .Names = c("name",
"type", "loops", "p")), structure(list(names = c("BE", "BG",
"CZ", "DK", "DE", "EE", "IE", "EL", "ES", "FR", "IT", "CY",
"LV", "LT", "LU", "HU", "MT", "NL", "AT", "PL", "PT", "RO",
"SI", "SK", "FI", "SE", "UK", "HR")), .Names = "names"),
structure(list(weight = c(16.3104184714332, 30.7293642773093,
1.51651715633125, 46.5498389394643, 117.400128622748,
9116.96298628885, 1101.62292417749, 12.3449820775232,
158.642126867059, 1.81516408646891, 760.982895191571,
1011.05564809154, 31677.1638228568, 189.344972592696,
13.1424122833109, 66.5517926330624, 12.0922729908787,
4.58139821481045, 90.1042955878783, 3.71064650396136,
90.4281997569113, 158.365362288124, 46.6972207187443,
30.3317841283049, 7.78960961626955, 1.95560245248322,
697.72330148994, 58.3866681241721, 660.780592399842,
1.30621866244087, 90.6566452600129, 411.372413931909,
50.4594917412583, 30.747955195432, 5.22931377396076,
26225.4597289607, 518.792688616195, 12998.9635603034,
69.4202716300456, 8.21344218621299, 0.582827301648975,
8.09867531337074, 107.652085596974, 1142.61073604216,
116.541846414264, 1.62280326041373, 405.914678311703,
0.386623934247936, 18.218603738158, 21.1501080346551,
0.158066706508139, 862.791072359265, 8.42653129127126,
20.0432021225717)), .Names = "weight"))), class = "igraph")
g <- upgrade_graph(g)
## l <- layout.fruchterman.reingold(g)
l <- layout_with_dh(g)
l <- layout.norm(l, -1,1, -1,1)
l <- norm_coords(layout.graphopt(g), xmin=-2, xmax=2, ymin=-2, ymax=2)
set.seed(1234)
pdf("toy-network.pdf")
plot(g, layout=l,
## rescale=F, ylim=c(0.4,2.6), xlim=c(-3.4, -1.7),
## vertex.label.dist=0.5,
vertex.color=adjustcolor("blue", .5),
vertex.frame.color="#ff000033",
edge.color="#55555533",
vertex.label=V(g)$names,
vertex.label.color="black",
vertex.size=log(strength(g))*2,
edge.width= log(1+E(g)$weight)
, edge.curved=0.3,
edge.label=NA )
dev.off()
_______________________________________________
igraph-help mailing list
https://lists.nongnu.org/mailman/listinfo/igraph-help
_______________________________________________
igraph-help mailing list
https://lists.nongnu.org/mailman/listinfo/igraph-help