how to create a minimun spanning tree with R
I'm trying to create a minimum spanning tree with R. My problem is that I have my distances stored in a txt as follow:
PID1 PID2 distance
1 4 0.002
2 5 0.004
3 6 0.003
Does there exist a way in which I can use my txt as matrix? I tried to transform it but I couldn't. Thank you everyone in advance
My code:
dat=read.table("distances.txt", dec='.', header=T)
Matrix<- as.matrix(dat)
mstree <-mst(Matrix)
Error en `[<-`(`*tmp*`, j, index.i, value = 139127) :
2 answers
In R, ?mst will display the help of the function. If you read the help of mst(), and then follow to the help of dist(), you will find out the particular matrix format the mst() function expects:
P1 P2 ... Pn
P1 0 0.002 d(1,n)
P2 0.002 0 d(2,n)
...
Pn d(1,n) d(2,n) 0
You are mixing the R data of type matrix with the particular matrix mst() expects. You will have a bit more of work to convert your data to a matrix of distances.
#load package
library(reshape2) #dcast() #reformatting table
library(fossil) #dino.mst() #mst function
#create a data frame with coordinates x and y and distance
dat <- data.frame(x=c(1,1,1,1,2,2,2,3,3,4),
y=c(1,2,3,4,2,3,4,3,4,4),
distance=c(0,0.2,0.5,0.1,0,0.5,0.8,0,0.1,0))
#convert to wide format
x <- dcast(dat,y~x,value.var="distance")
#remove first column
x <- x[,-1]
#convert to a dist object
x <- as.dist(x)
#this dist object can be used in any function that accepts a dist object.
#calculate mst
mstobj <- dino.mst(x)
Log in to answer this question.
More posts like this
This is not a bioinformatics question but an R programming one. You may have better luck getting an answer on StackOverflow.
First issue is that your data file is not in matrix form. as.matrix() is not going to do any rearranging, it just converts internal data types. Second, mstree() (from the spdep package) doesn't take a matrix as argument. Read the docs.
Finally, for graph-related work in R, I suggest using the igraph package.
I guess Lila M is using
mst()from {ape} package, and it may take a matrix as argument.