Automated determination of distribution groupings – A StackOverflow collaboration

May 18, 2014
By

(This article was first published on me nugget, and kindly contributed to R-bloggers)

For those of you not familiar with StackOverflow (SO), it's a coder's help forum on the StackExchange website. It's one of the best resources for R-coding tips that I know of, due entirely to the community of users that routinely give expert advise (assuming you show that you have done your homework and provide a clear question and a reproducible example). It's hard to believe that users spend time to offer this help for nothing more than virtual reputation points. I think a lot of coders are probably puzzle fanatics at heart, and enjoy the challenge of a given problem, but I'm nevertheless amazed by the depth of some of the R-related answers. The following is a short example of the value of this community (via SO), which helped me find a solution to a tricky problem.

I have used figures like the one above (left) in my work at various times. It present various distributions in the form of a boxplot, and uses differing labels (in this case, the lowercase letters) to denote significant differences; i.e. levels sharing a label are not significantly different. This type of presentation is common when showing changes in organism condition indices over time (e.g. Figs 3 & 4, Bullseye puffer fish in Mexico).

In the example above, a Kruskal-Wallis rank sum test is used to test differences across all levels, followed by pairwise Mann-Whitney rank tests. The result is a matrix of p-values showing significant differences in distribution. So far so good, but it's not always clear how the grouping relationships should be labelled. In this relatively simple example, the tricky part is that level 1 should be grouped with 3 and 5, but 3 and 5 should not be grouped; Therefore, two labeling codes should be designated, with level 1 sharing both. I have wondered, for some time, if there might be some way to do this in an automated fashion using an algorithm. After many attempts on my own, I finally decided to post a question to SO.

So, my first question "Algorithm for automating pairwise significance grouping labels in R" led me to the concept of the "clique cover problem", and "graph theory" in general, via SO user "David Eisenstat". While I didn't completely understand his recommendation at first, it got me pointed in the right direction - I ultimately found the R package igraph for analyzing and plotting these types of problems.

The next questions were a bit more technical. I figured out that I could return the "cliques" of my grouping relationships network using the cliques function of the igraph package, but my original attempt was giving me a list all relationships in my matrix. It was obvious to me that I would need to identify groupings where all levels were fully connected (i.e. each node in the clique connects to all others). So, my next question "How to identify fully connected node clusters with igraph [in R]" got me a tip from SO user "majom", who showed me that these fully connected cliques could be identified by first reordering the starting nodes in my list of connections (before use in the graph.data.frame function), and then subjecting the resulting igraph object to the function maximal.cliques. So, the first suggestions from David were right on, even though they didn't include code. The result shows nicely all those groupings in the above example (right plot) with fully connected cliques [i.e. (1, 3), (1, 5), (2), (4, 6), (7)].

The final piece of the puzzle was more cosmetic - "How to order a list of vectors based on the order of values contained within [in R]". A bit vague, I know, but what I was trying to do was to label groups in a progressive way so that earlier levels received their labels first. I think this leads to more legible labeling, especially when levels represent some process of progression. At the time of this posting, I have received a single negative (-1) vote on this question... This may have to do with the clarity of the question - I seem to have confused some of the respondents based on follow up comments for clarification - or, maybe someone thought I hadn't shown enough effort on my own. There's no way to know without an accompanying comment. In any case, I got a robust approach from SO user "MrFlick", and I can safely say that I would never have come up with such an eloquent solution on my own.

In all, this solution seems to work great. I have tried it out on larger problems involving more levels and it appears to give correct results. Here is an example with 20 levels (a problem that would have been an amazing headache to do manually):
Any comments are welcome. There might be other ways of doing this (clustering?), but searching for similar methods seems to be limited by my ability to articulate the problem. Who would have thought this was an example of a "clique cover problem"? Thanks again to all those that provided help on SO!

Code to reproduce the example:

###Required packagelibrary(igraph) ### Synthetic data setset.seed(1)n <- 7n2 <- 100mu <- cumsum(runif(n, min=-3, max=3))sigma <- runif(n, min=1, max=3) dat <- vector(mode="list", n)for(i in seq(dat)){ dat[[i]] <- rnorm(n2, mean=mu[i], sd=sigma[i])} df <- data.frame(group=as.factor(rep(seq(n), each=n2)), y=unlist(dat)) ### Plotpng("boxplot_levels_nolabels.png", width=6, height=4, units="in", res=400, type="cairo")par(mar=c(4,4,1,1))boxplot(y ~ group, df, notch=TRUE, outline=FALSE) # boxplot of factor level distributionsmtext("Levels", side=1, line=2.5)mtext("y", side=2, line=2.5)dev.off() ### Test for significant differences in factor level distributionskruskal.test(y ~ group, df) # Significant differences as determined by Kruskal-Wallis rank sum testpairwise.wilcox.test(df$y, df$g) # Pairwise Wilcoxon (or "Mann-Whitney") rank sum tests between all factor level combinations ### Labeling of factor level groupingsmw <- pairwise.wilcox.test(df$y, df$g) # Create matrix showing factor levels that should be groupedg <- as.matrix(mw$p.value > 0.05) # TRUE signifies that pairs of not significantly different at the p < 0.05 levelg <- cbind(rbind(NA, g), NA) # make squareg <- replace(g, is.na(g), FALSE) # replace NAs with FALSEg <- g + t(g) # not necessary, but make matrix symmetricdiag(g) <- 1 # diagonal equals 1rownames(g) <- 1:n # change row namescolnames(g) <- 1:n # change column namesg # resulting matrix # Re-arrange data into an "edge list" for use in igraph (i.e. which groups are "connected") - Solution from "David Eisenstat" ()same <- which(g==1)g2 <- data.frame(N1=((same-1) %% n) + 1, N2=((same-1) %/% n) + 1)g2 <- g2[order(g2[[1]]),] # Get rid of loops and ensure right naming of verticesg3 <- simplify(graph.data.frame(g2,directed = FALSE))get.data.frame(g3) # view connections # Plot igraphpng("igraph_level_groupings.png", width=5, height=5, units="in", res=400, type="cairo")par(mar=c(3,1,1,1))V(g3)$color <- 8V(g3)$label.color <- 1V(g3)$size <- 20plot(g3) # plot all nodes are connectionsbox()mtext("Linked levels are not significantly different \n(Mann-Whitney)", side=1, line=1)dev.off() # Calcuate the maximal cliques - these are groupings where every node is connected to all otherscliq <- maximal.cliques(g3) # Solution from "majom" () # Reorder by level order - Solution from "MrFlick" ()ml<-max(sapply(cliq, length))reord <- do.call(order, data.frame(    do.call(rbind,         lapply(cliq, function(x) c(sort(x), rep.int(0, ml-length(x))))    )))cliq <- cliq[reord]cliq # Generate labels to  factor levelslab.txt <- vector(mode="list", n) # empty listlab <- letters[seq(cliq)] # clique labelsfor(i in seq(cliq)){ # loop to concatenate clique labels for(j in cliq[[i]]){  lab.txt[[j]] <- paste0(lab.txt[[j]], lab[i]) }} # Boxplot with facor level grouping labelspng("boxplot_levels_withlabels.png", width=6, height=4, units="in", res=400, type="cairo")par(mar=c(4,4,1,1))ylim <- range(df$y) + c(0,2)bp <- boxplot(y ~ group, df, notch=TRUE, ylim=ylim, outline=FALSE) # boxplot of factor level distributionstext(x=1:n, y=bp$stats[4,], labels=lab.txt, col=1, cex=1, font=2, adj=c(-0.2, -0.2))mtext("Levels", side=1, line=2.5)mtext("y", side=2, line=2.5)dev.off()
Created by Pretty R at inside-R.org