krashkov / steinernet Goto Github PK

@krashkov @afshinsadeghi: This package does not terminate its checks when run with the release candidate for igraph 2.0.0.

Is this a problem in your package? Should we be doing things differently?

Install the 2.0.0 release candidate using pak::pak("igraph/rigraph") .

We're planning to release igraph 2.0.0 on January 23. My sincere apologies for the short notice, I only realized today that I should have warned about the problem much earlier. Happy to postpone if it turns out that the problem is on our end. Thank you for your help!

Tracker: igraph/rigraph#989.

Hello!

I installed SteinerNet according to the documentation "https://www.rdocumentation.org/packages/SteinerNet/versions/2.0/topics/steinertree".
However, when I tried to run Example 1 in this link. I always got the error message "Error in V(g) : Not a graph object".

Could you please tell me how to solve this problem?

Thank you!

Hello all,

Can anyone tell me how to retain the name of vertices in steinertree package (R)? Below is the code I used. The final object SP (igraph object) is the one that is showing node names as 1,2,3..

library(igraph)
    library(SteinerNet)

    a <- read.table("interactions.txt", sep="\t", quote="")
    head(a)
                   V1
    1 ENSP00000222390
    2 ENSP00000391826
    3 ENSP00000222390
    4 ENSP00000391669
    5 ENSP00000222390
    6 ENSP00000384675

    el <- matrix(as.vector(a$V1), nc=2, byrow=T) 

    g1 =graph_from_edgelist(el, directed=F)

    ter_list= c("ENSP00000306512","ENSP00000340019","ENSP00000361125","ENSP00000367207","ENSP00000384273","ENSP00000388107","ENSP00000400175","ENSP00000401303")
    SP=steinertree(type ="SP",terminals = ter_list, graph = g1)

SP
[[1]]
IGRAPH 6a21ce1 UN-- 139 1600 -- 
+ attr: name (v/c), realname (v/c), color (v/c)
+ edges from 6a21ce1 (vertex names):
 [1] 1 --2  1 --3  1 --4  1 --5  1 --6  1 --7  1 --8  1 --9  1 --10 1 --11
[11] 1 --12 1 --13 1 --14 1 --15 1 --16 1 --17 1 --18 1 --19 1 --20 1 --21
[21] 1 --22 1 --23 1 --24 1 --25 1 --26 1 --27 1 --28 1 --29 1 --30 1 --31
[31] 1 --32 1 --33 34--35 34--36 34--37 34--38 9 --34 34--39 20--34 34--40
[41] 25--34 27--34 34--41 34--42 34--43 34--44 35--45 45--46 45--47 45--48
[51] 38--45 45--49 39--45 40--45 45--50 30--51 3 --30 4 --30 30--52 30--53
[61] 30--54 5 --30 30--48 30--55 8 --30 30--56 9 --30 10--30 30--57 11--30
[71] 15--30 30--58 20--30 21--30 30--59 25--30 30--60 27--30 28--30 3 --61
+ ... omitted several edges

[[2]]
IGRAPH 886b59f UN-- 12 11 -- 
+ attr: name (v/c), realname (v/c), color (v/c)
+ edges from 886b59f (vertex names):
 [1] 5 --8  5 --19 5 --21 5 --50 35--50 5 --51 21--53 5 --61 8 --69 53--73
[11] 5 --84

Please help me out. Thanks.

Hi,
I believe there is an issue with the merging of non-disjoint subtrees in the steinertree3 function in steinertree.R (lines 261 and following):
I encountered a case where there is no overlap between my trees with i=1 and j=2. So j is increased to 3. In this case there was an overlap between the i=1 and j=3 trees. When j reaches the end of the subtree list, i is increased. The problem is that the tree at i=1 is not the same anymore than the one that has been tested against j=2 (because it has been merged with other trees), so the new i=1 and j=2 trees may actually overlap. In my case, this lead to an empty list when calculating 'nsubtrees', which cased an error.
A solution may be to not increase i in a loop where something has been merged into the corresponding tree. My code is too complex to share here, but I hope the explanation makes sense.