1 ############FIXME##################
2 #PlotDendroComp <- function(chd,filename,reso) {
3 # jpeg(filename,res=reso)
5 # plot(chd,which.plots=2, hang=-1)
9 #PlotDendroHori <- function(dendrocutupper,filename,reso) {
10 # jpeg(filename,res=reso)
12 # nP <- list(col=3:2, cex=c(0.5, 0.75), pch= 21:22, bg= c('light blue', 'pink'),lab.cex = 0.75, lab.col = 'tomato')
13 # plot(dendrocutupper,nodePar= nP, edgePar = list(col='gray', lwd=2),horiz=TRUE, center=FALSE)
17 PlotDendroCut <- function(chd,filename,reso,clusternb) {
18 h.chd <- as.hclust(chd)
19 memb <- cutree(h.chd, k = clusternb)
21 for(k in 1:clusternb){
22 cent <- rbind(cent, k)
24 h.chd1 <- hclust(dist(cent)^2, method = 'cen', members = table(memb))
25 h.chd1$labels <- sprintf('CL %02d',1:clusternb)
26 nP <- list(col=3:2, cex=c(2.0, 0.75), pch= 21:22, bg= c('light blue', 'pink'),lab.cex = 0.75, lab.col = 'tomato')
27 jpeg(filename,res=reso)
29 plot(h.chd1, nodePar= nP, edgePar = list(col='gray', lwd=2), horiz=TRUE, center=TRUE, hang= -1)
33 #PlotAfc<- function(afc, filename, width=800, height=800, quality=100, reso=200, toplot=c('all','all'), PARCEX=PARCEX) {
34 # if (Sys.info()["sysname"]=='Darwin') {
36 # height<-height/74.97
37 # quartz(file=filename,type='jpeg',width=width,height=height)
39 # jpeg(filename,width=width,height=height,quality=quality,res=reso)
42 # plot(afc,what=toplot,labels=c(1,1),contrib=c('absolute','relative'))
46 PlotAfc2dCoul<- function(afc,chisqrtable,filename, what='coord',col=FALSE, axetoplot=c(1,2), deb=0,fin=0, width=900, height=900, quality=100, reso=200, parcex=PARCEX, xlab = NULL, ylab = NULL, xmin=NULL, xmax=NULL, ymin=NULL, ymax=NULL, active = TRUE) {
48 if (what == 'coord') {
49 rowcoord <- as.matrix(afc$colcoord)
51 rowcoord <- as.matrix(afc$colcrl)
54 if (what == 'coord') {
55 rowcoord <- as.matrix(afc$rowcoord)
57 rowcoord <- as.matrix(afc$rowcrl)
63 rownames(rowcoord) <- afc$colnames
65 rownames(rowcoord) <- afc$rownames
66 rowcoord <- as.matrix(rowcoord[deb:fin,])
67 chitable<- as.matrix(chisqrtable[deb:fin,])
68 #row_keep <- select_point_nb(chitable,15)
70 if (ncol(rowcoord) == 1) {
71 rowcoord <- t(rowcoord)
73 clnb <- ncol(chisqrtable)
76 classes <- as.matrix(apply(chitable,1,which.max))
77 cex.par <- norm.vec(apply(chitable,1,max), 0.8,3)
78 row.keep <- select.chi.classe(chitable, 80, active=active)
79 rowcoord <- rowcoord[row.keep,]
80 classes <- classes[row.keep]
81 cex.par <- cex.par[row.keep]
84 cex.par <- rep(1,clnb)
88 xminmax <- c(min(table.in[,1], na.rm = TRUE) + ((max(cex.par)/10) * min(table.in[,1], na.rm = TRUE)), max(table.in[,1], na.rm = TRUE) + ((max(cex.par)/10) * max(table.in[,1], na.rm = TRUE)))
91 yminmax <- c(min(table.in[,2], na.rm = TRUE) + ((max(cex.par)/10) * min(table.in[,2], na.rm = TRUE)), max(table.in[,2], na.rm = TRUE) + ((max(cex.par)/10) * max(table.in[,2], na.rm = TRUE)))
95 #ntabtot <- cbind(rowcoord, classes)
96 #if (!col) ntabtot <- ntabtot[row_keep,]
97 xlab <- paste('facteur ', x, ' -')
98 ylab <- paste('facteur ', y, ' -')
99 xlab <- paste(xlab,round(afc_table$facteur[x,2],2),sep = ' ')
100 xlab <- paste(xlab,' %%',sep = '')
101 ylab <- paste(ylab,round(afc_table$facteur[y,2],2),sep = ' ')
102 ylab <- paste(ylab,' %%',sep = '')
104 open_file_graph(filename, width = width, height = height)
106 table.in <- rowcoord[order(cex.par, decreasing = TRUE),]
107 classes <- classes[order(cex.par, decreasing = TRUE)]
108 cex.par <- cex.par[order(cex.par, decreasing = TRUE)]
109 table.out <- stopoverlap(table.in, cex.par=cex.par, xlim = c(xmin,xmax), ylim = c(ymin,ymax))
110 table.in <- table.out$toplot
111 notplot <- table.out$notplot
112 if (! is.null(notplot)) {
113 write.csv2(notplot, file = paste(filename, '_notplotted.csv', sep = ''))
115 classes <- classes[table.in[,4]]
116 cex.par <- cex.par[table.in[,4]]
117 make_afc_graph(table.in, classes, clnb, xlab, ylab, cex.txt = cex.par, xminmax=c(xmin,xmax), yminmax=c(ymin,ymax))
118 xyminmax <- list(yminmax = c(ymin,ymax), xminmax = c(xmin,xmax))
120 #plot(rowcoord[,x],rowcoord[,y], pch='', xlab = xlab, ylab = ylab)
123 # ntab <- subset(ntabtot,ntabtot[,ncol(ntabtot)] == i)
124 # if (nrow(ntab) != 0)
125 # text(ntab[,x],ntab[,y],rownames(ntab),col=rainbow(clnb)[i])
130 filename.to.svg <- function(filename) {
131 filename <- gsub('.png', '.svg', filename)
135 open_file_graph <- function (filename, width=800, height = 800, quality = 100, svg = FALSE) {
136 if (Sys.info()["sysname"] == 'Darwin') {
138 height <- height/74.97
140 quartz(file = filename, type = 'png', width = width, height = height)
142 svg(filename.to.svg(filename), width=width, height=height)
146 svg(filename.to.svg(filename), width=width/74.97, height=height/74.97)
148 png(filename, width=width, height=height)#, quality = quality)
153 #################################################@@
155 overlap <- function(x1, y1, sw1, sh1, boxes) {
156 use.r.layout <- FALSE
158 return(.overlap(x1,y1,sw1,sh1,boxes))
160 if (length(boxes) == 0)
162 for (i in c(last,1:length(boxes))) {
169 overlap <- x1 + sw1 > x2-s
171 overlap <- x2 + sw2 > x1-s
174 overlap <- overlap && (y1 + sh1 > y2-s)
176 overlap <- overlap && (y2 + sh2 > y1-s)
185 .overlap <- function(x11,y11,sw11,sh11,boxes1){
186 if (as.character(packageVersion('wordcloud')) >= '2.6') {
187 .Call("_wordcloud_is_overlap", x11,y11,sw11,sh11,boxes1)
189 .Call("is_overlap",x11,y11,sw11,sh11,boxes1)
192 ########################################################
193 stopoverlap <- function(x, cex.par = NULL, xlim = NULL, ylim = NULL) {
205 plot(x[,1],x[,2], pch='', xlim = xlim, ylim = ylim)
208 if (is.null(cex.par)) {
209 size <- rep(0.9, nrow(x))
213 #cols <- rainbow(clnb)
215 for (i in 1:nrow(x)) {
216 rotWord <- runif(1)<rot.per
218 theta <- runif(1,0,2*pi)
221 wid <- strwidth(words[i],cex=size[i])
222 ht <- strheight(words[i],cex=size[i])
225 if(!overlap(x1-.5*wid,y1-.5*ht,wid,ht, boxes)) { #&&
226 toplot <- rbind(toplot, c(x1, y1, size[i], i))
227 #text(x1,y1,words[i],cex=size[i],offset=0,srt=rotWord*90,
228 # col=cols[classes[i]])
229 boxes[[length(boxes)+1]] <- c(x1-.5*wid,y1-.5*ht,wid,ht)
233 #print(paste(words[i], "could not be fit on page. It will not be plotted."))
234 notplot <- rbind(notplot,c(words[i], x[i,1], x[i,2], size[i], i))
237 theta <- theta+thetaStep
238 r <- r + rStep*thetaStep/(2*pi)
239 x1 <- x[i,1]+r*cos(theta)
240 y1 <- x[i,2]+r*sin(theta)
244 nbnot <- nrow(notplot)
245 print(paste(nbnot, ' not plotted'))
246 row.names(toplot) <- words[toplot[,4]]
247 return(list(toplot = toplot, notplot = notplot))
249 ###############################################################################
251 getwordcloudcoord <- function(words,freq,scale=c(4,.5),min.freq=3,max.words=Inf,random.order=TRUE,random.color=FALSE,
252 rot.per=.1,colors="black",ordered.colors=FALSE,use.r.layout=FALSE,fixed.asp=TRUE,...) {
256 overlap <- function(x1, y1, sw1, sh1) {
258 return(.overlap(x1,y1,sw1,sh1,boxes))
260 if (length(boxes) == 0)
262 for (i in c(last,1:length(boxes))) {
269 overlap <- x1 + sw1 > x2-s
271 overlap <- x2 + sw2 > x1-s
274 overlap <- overlap && (y1 + sh1 > y2-s)
276 overlap <- overlap && (y2 + sh2 > y1-s)
285 ord <- rank(-freq, ties.method = "random")
286 words <- words[ord<=max.words]
287 freq <- freq[ord<=max.words]
290 ord <- order(freq,decreasing=TRUE)
293 words <- words[freq>=min.freq]
294 freq <- freq[freq>=min.freq]
295 if (ordered.colors) {
296 colors <- colors[ord][freq>=min.freq]
303 normedFreq <- freq/max(freq)
304 size <- (scale[1]-scale[2])*normedFreq + scale[2]
309 for(i in 1:length(words)){
310 rotWord <- runif(1)<rot.per
312 theta <- runif(1,0,2*pi)
315 wid <- strwidth(words[i],cex=size[i],...)
316 ht <- strheight(words[i],cex=size[i],...)
318 if(grepl(tails,words[i]))
327 if(!overlap(x1-.5*wid,y1-.5*ht,wid,ht) &&
328 x1-.5*wid>0 && y1-.5*ht>0 &&
329 x1+.5*wid<1 && y1+.5*ht<1){
330 toplot <- rbind(toplot, c(x1,y1,size[i], i))
331 boxes[[length(boxes)+1]] <- c(x1-.5*wid,y1-.5*ht,wid,ht)
335 warning(paste(words[i],
336 "could not be fit on page. It will not be plotted."))
339 theta <- theta+thetaStep
340 r <- r + rStep*thetaStep/(2*pi)
341 x1 <- .5+r*cos(theta)
342 y1 <- .5+r*sin(theta)
346 toplot <- cbind(toplot,norm.vec(freq[toplot[,4]], 1, 50))
347 row.names(toplot) <- words[toplot[,4]]
348 toplot <- toplot[,-4]
352 new_tree_tot <- function(chd) {
354 m <- matrix(0, ncol=2)
355 for (val in 1:length(lf)) {
356 if (! is.null(lf[[val]])) {
357 print(c(val,lf[[val]][1]))
358 m <- rbind(m, c(val,lf[[val]][1]))
359 m <- rbind(m, c(val,lf[[val]][2]))
365 make_tree_tot <- function (chd) {
369 for (i in 1:length(lf)) {
370 if (!is.null(lf[[i]])) {
371 clus<-gsub(paste('a',i,'a',sep=''),paste('(','a',lf[[i]][1],'a',',','a',lf[[i]][2],'a',')',sep=''),clus)
375 clus <- gsub('a','',clus)
376 tree.cl <- read.tree(text = clus)
377 res<-list(tree.cl = tree.cl, dendro_tuple = dendro_tuple)
381 make_dendro_cut_tuple <- function(dendro_in, coordok, classeuce, x, nbt = 9) {
385 for (cl in coordok[,x]) {
387 fcl<-fille(cl,classeuce)
389 dendro <- gsub(paste('a',fi,'a',sep=''),paste('b',i,'b',sep=''),dendro)
392 clnb <- nrow(coordok)
394 for (i in 1:(tcl + 1)) {
395 dendro <- gsub(paste('a',i,'a',sep=''),paste('b',0,'b',sep=''),dendro)
397 dendro <- gsub('b','',dendro)
398 dendro <- gsub('a','',dendro)
399 dendro_tot_cl <- read.tree(text = dendro)
403 dendro <- gsub(paste('\\(',cl,',',cl,'\\)',sep=''),cl,dendro)
407 dendro <- gsub(paste('\\(',0,',',0,'\\)',sep=''),0,dendro)
409 dendro <- gsub(paste('\\(',0,',',cl,'\\)',sep=''),cl,dendro)
410 dendro <- gsub(paste('\\(',cl,',',0,'\\)',sep=''),cl,dendro)
414 tree.cl <- read.tree(text = dendro)
415 lab <- tree.cl$tip.label
417 tovire <- which(lab == "0")
418 tree.cl <- drop.tip(tree.cl, tip = tovire)
420 res <- list(tree.cl = tree.cl, dendro_tuple_cut = dendro, dendro_tot_cl = dendro_tot_cl)
424 select_point_nb <- function(tablechi, nb) {
425 chimax<-as.matrix(apply(tablechi,1,max))
426 chimax<-cbind(chimax,1:nrow(tablechi))
427 order_chi<-as.matrix(chimax[order(chimax[,1],decreasing = TRUE),])
428 row_keep <- order_chi[,2][1:nb]
432 select_point_chi <- function(tablechi, chi_limit) {
433 chimax<-as.matrix(apply(tablechi,1,max))
434 row_keep <- which(chimax >= chi_limit)
438 select.chi.classe <- function(tablechi, nb, active = TRUE) {
440 if (active & !is.null(debsup)) {
441 tablechi <- tablechi[1:(debsup-1),]
443 if (nb > nrow(tablechi)) {
446 for (i in 1:ncol(tablechi)) {
447 rowkeep <- append(rowkeep,order(tablechi[,i], decreasing = TRUE)[1:nb])
449 rowkeep <- unique(rowkeep)
453 select.chi.classe.et <- function(tablechi, nb){
455 if (!is.null(debet)) {
456 ntablechi <- tablechi[debet:nrow(tablechi),]
458 if (nb > nrow(ntablechi)) {
459 nb <- nrow(ntablechi)
461 for (i in 1:ncol(ntablechi)) {
462 rowkeep <- append(rowkeep,order(ntablechi[,i], decreasing = TRUE)[1:nb])
464 rowkeep <- unique(rowkeep)
469 summary.ca.dm <- function(object, scree = TRUE, ...){
475 if (nd > length(obj$sv)) nd <- length(obj$sv)
477 # principal coordinates:
479 I <- dim(obj$rowcoord)[1] ; J <- dim(obj$colcoord)[1]
480 svF <- matrix(rep(obj$sv[1:K], I), I, K, byrow = TRUE)
481 svG <- matrix(rep(obj$sv[1:K], J), J, K, byrow = TRUE)
482 rpc <- obj$rowcoord[,1:K] * svF
483 cpc <- obj$colcoord[,1:K] * svG
486 r.names <- obj$rownames
488 if (!is.na(sr[1])) r.names[sr] <- paste("(*)", r.names[sr], sep = "")
489 r.mass <- obj$rowmass
490 r.inr <- obj$rowinertia / sum(obj$rowinertia, na.rm = TRUE)
491 r.COR <- matrix(NA, nrow = length(r.names), ncol = nd)
492 colnames(r.COR) <- paste('COR -facteur', 1:nd, sep=' ')
493 r.CTR <- matrix(NA, nrow = length(r.names), ncol = nd)
494 colnames(r.CTR) <- paste('CTR -facteur', 1:nd, sep=' ')
496 r.COR[,i] <- obj$rowmass * rpc[,i]^2 / obj$rowinertia
497 r.CTR[,i] <- obj$rowmass * rpc[,i]^2 / obj$sv[i]^2
499 # cor and quality for supplementary rows
500 if (length(obj$rowsup) > 0){
503 r.COR[i0,i] <- obj$rowmass[i0] * rpc[i0,i]^2
509 c.names <- obj$colnames
511 if (!is.na(sc[1])) c.names[sc] <- paste("(*)", c.names[sc], sep = "")
512 c.mass <- obj$colmass
513 c.inr <- obj$colinertia / sum(obj$colinertia, na.rm = TRUE)
514 c.COR <- matrix(NA, nrow = length(c.names), ncol = nd)
515 colnames(c.COR) <- paste('COR -facteur', 1:nd, sep=' ')
516 c.CTR <- matrix(NA, nrow = length(c.names), ncol = nd)
517 colnames(c.CTR) <- paste('CTR -facteur', 1:nd, sep=' ')
520 c.COR[,i] <- obj$colmass * cpc[,i]^2 / obj$colinertia
521 c.CTR[,i] <- obj$colmass * cpc[,i]^2 / obj$sv[i]^2
523 if (length(obj$colsup) > 0){
526 c.COR[i0,i] <- obj$colmass[i0] * cpc[i0,i]^2
534 values2 <- 100*(obj$sv^2)/sum(obj$sv^2)
535 values3 <- cumsum(100*(obj$sv^2)/sum(obj$sv^2))
536 scree.out <- cbind(values, values2, values3)
545 obj$facteur <- scree.out
549 create_afc_table <- function(x) {
551 facteur.table <- as.matrix(x$facteur)
552 nd <- ncol(x$colcoord)
553 rownames(facteur.table) <- paste('facteur',1:nrow(facteur.table),sep = ' ')
554 colnames(facteur.table) <- c('Valeurs propres', 'Pourcentages', 'Pourcentage cumules')
555 ligne.table <- as.matrix(x$rowcoord)
556 rownames(ligne.table) <- x$rownames
557 colnames(ligne.table) <- paste('Coord. facteur', 1:nd, sep=' ')
558 tmp <- as.matrix(x$rowcrl)
559 colnames(tmp) <- paste('Corr. facteur', 1:nd, sep=' ')
560 ligne.table <- cbind(ligne.table,tmp)
561 ligne.table <- cbind(ligne.table, x$r.COR)
562 ligne.table <- cbind(ligne.table, x$r.CTR)
563 ligne.table <- cbind(ligne.table, mass = x$rowmass)
564 ligne.table <- cbind(ligne.table, chi.distance = x$rowdist)
565 ligne.table <- cbind(ligne.table, inertie = x$rowinertia)
566 colonne.table <- x$colcoord
567 rownames(colonne.table) <- paste('classe', 1:(nrow(colonne.table)),sep=' ')
568 colnames(colonne.table) <- paste('Coord. facteur', 1:nd, sep=' ')
569 tmp <- as.matrix(x$colcrl)
570 colnames(tmp) <- paste('Corr. facteur', 1:nd, sep=' ')
571 colonne.table <- cbind(colonne.table, tmp)
572 colonne.table <- cbind(colonne.table, x$c.COR)
573 colonne.table <- cbind(colonne.table, x$c.CTR)
574 colonne.table <- cbind(colonne.table, mass = x$colmass)
575 colonne.table <- cbind(colonne.table, chi.distance = x$coldist)
576 colonne.table <- cbind(colonne.table, inertie = x$colinertia)
577 res <- list(facteur = facteur.table, ligne = ligne.table, colonne = colonne.table)
581 is.yellow <- function(my.color) {
582 if ((my.color[1] > 200) & (my.color[2] > 200) & (my.color[3] < 20)) {
589 del.yellow <- function(colors) {
590 rgbs <- col2rgb(colors)
591 tochange <- apply(rgbs, 2, is.yellow)
592 tochange <- which(tochange)
593 if (length(tochange)) {
594 gr.col <- grey.colors(length(tochange), start = 0.5, end = 0.8)
597 for (val in tochange) {
598 colors[val] <- gr.col[compt]
604 make_afc_graph <- function(toplot, classes, clnb, xlab, ylab, cex.txt = NULL, leg = FALSE, cmd = FALSE, black = FALSE, xminmax=NULL, yminmax=NULL, color=NULL) {
606 rain <- rainbow(clnb)
609 #for (my.color in rain) {
610 # my.color <- col2rgb(my.color)
611 # if ((my.color[1] > 200) & (my.color[2] > 200) & (my.color[3] < 20)) {
612 # tochange <- append(tochange, compt)
616 #if (!is.null(tochange)) {
617 # gr.col <- grey.colors(length(tochange))
619 # for (val in tochange) {
620 # rain[val] <- gr.col[compt]
624 rain <- del.yellow(rain)
625 cl.color <- rain[classes]
629 if (!is.null(color)) {
632 plot(toplot[,1],toplot[,2], pch='', xlab = xlab, ylab = ylab, xlim=xminmax, ylim = yminmax)
633 abline(h=0, v=0, lty = 'dashed')
634 if (is.null(cex.txt))
635 text(toplot[,1],toplot[,2],rownames(toplot),col=cl.color, offset=0)
638 #textplot(toplot[,1],toplot[,2],rownames(toplot),col=cl.color, cex = cex.txt, xlim=xminmax, ylim = yminmax)
639 text(toplot[,1],toplot[,2],rownames(toplot),col=cl.color, cex = cex.txt, offset=0)
646 plot.dendro.prof <- function(tree, classes, chisqtable, nbbycl = 60, type.dendro = "phylogram", from.cmd = FALSE, bw = FALSE, lab = NULL) {
649 classes<-classes[classes!=0]
650 classes<-as.factor(classes)
651 sum.cl<-as.matrix(summary(classes, maxsum=1000000))
652 sum.cl<-(sum.cl/colSums(sum.cl)*100)
653 sum.cl<-round(sum.cl,2)
654 sum.cl<-cbind(sum.cl,as.matrix(100-sum.cl[,1]))
656 tree.order<- as.numeric(tree$tip.label)
658 row.keep <- select.chi.classe(chisqtable, nbbycl)
659 #et.keep <- select.chi.classe.et(chisqtable, 10)
660 #print(chistable[et.keep,])
661 toplot <- chisqtable[row.keep,]
663 for (classe in 1:length(sum.cl)) {
664 ntoplot <- toplot[,classe]
665 names(ntoplot) <- rownames(toplot)
666 ntoplot <- ntoplot[order(ntoplot, decreasing = TRUE)]
667 ntoplot <- round(ntoplot, 0)
668 if (length(toplot) > nbbycl) {
669 ntoplot <- ntoplot[1:nbbycl]
671 ntoplot <- ntoplot[which(ntoplot > 0)]
672 #ntoplot <- ntoplot[order(ntoplot)]
673 #ntoplot <- ifelse(length(ntoplot) > nbbycl, ntoplot[1:nbbycl], ntoplot)
674 lclasses[[classe]] <- ntoplot
676 vec.mat <- matrix(1, nrow = 3, ncol = length(sum.cl))
678 vec.mat[3,] <- 3:(length(sum.cl)+2)
679 layout(matrix(vec.mat, nrow=3, ncol=length(sum.cl)),heights=c(2,1,6))
681 col <- rainbow(length(sum.cl))
682 col <- del.yellow(col)
683 col <- col[as.numeric(tree$tip.label)]
684 colcloud <- rainbow(length(sum.cl))
685 colcloud <- del.yellow(colcloud)
687 label.ori<-tree$tip.label
689 tree$tip.label <- lab
691 tree$tip.label<-paste('classe ',tree$tip.label)
694 plot.phylo(tree,label.offset=0, tip.col=col, type=type.dendro, direction = 'downwards', srt=90, adj = 0.5, cex = 1.5, y.lim=c(-0.3,tree$Nnode))
696 d <- barplot(-sum.cl[tree.order], col=col, names.arg='', axes=FALSE, axisname=FALSE)
697 text(x=d, y=(-sum.cl[tree.order]+3), label=paste(round(sum.cl[tree.order],1),'%'), cex=1)
698 for (i in tree.order) {
699 par(mar=c(0,0,1,0),cex=0.7)
700 #wordcloud(names(lclasses[[i]]), lclasses[[i]], scale = c(1.5, 0.2), random.order=FALSE, colors = colcloud[i])
702 plot(0,0,pch='', axes = FALSE)
703 vcex <- norm.vec(lclasses[[i]], 2, 3)
704 for (j in 1:length(lclasses[[i]])) {
705 yval <- yval-(strheight( names(lclasses[[i]])[j],cex=vcex[j])+0.02)
706 text(-0.9, yval, names(lclasses[[i]])[j], cex = vcex[j], col = colcloud[i], adj=0)
715 plot.dendro.cloud <- function(tree, classes, chisqtable, nbbycl = 60, type.dendro = "phylogram", from.cmd = FALSE, bw = FALSE, lab = NULL) {
718 classes<-classes[classes!=0]
719 classes<-as.factor(classes)
720 sum.cl<-as.matrix(summary(classes, maxsum=1000000))
721 sum.cl<-(sum.cl/colSums(sum.cl)*100)
722 sum.cl<-round(sum.cl,2)
723 sum.cl<-cbind(sum.cl,as.matrix(100-sum.cl[,1]))
725 tree.order<- as.numeric(tree$tip.label)
727 row.keep <- select.chi.classe(chisqtable, nbbycl)
728 toplot <- chisqtable[row.keep,]
730 for (classe in 1:length(sum.cl)) {
731 ntoplot <- toplot[,classe]
732 names(ntoplot) <- rownames(toplot)
733 ntoplot <- ntoplot[order(ntoplot, decreasing = TRUE)]
734 ntoplot <- round(ntoplot, 0)
735 if (length(toplot) > nbbycl) {
736 ntoplot <- ntoplot[1:nbbycl]
738 ntoplot <- ntoplot[order(ntoplot)]
739 ntoplot <- ntoplot[which(ntoplot > 0)]
740 #ntoplot <- ifelse(length(ntoplot) > nbbycl, ntoplot[1:nbbycl], ntoplot)
741 lclasses[[classe]] <- ntoplot
743 for (i in 1:length(sum.cl)) vec.mat<-append(vec.mat,1)
745 for (i in 1:length(sum.cl)) {
746 vec.mat<-append(vec.mat,v)
749 layout(matrix(vec.mat,length(sum.cl),2),widths=c(1,2))
751 col <- rainbow(length(sum.cl))[as.numeric(tree$tip.label)]
752 colcloud <- rainbow(length(sum.cl))
755 label.ori<-tree$tip.label
757 tree$tip.label <- lab
759 tree$tip.label<-paste('classe ',tree$tip.label)
761 plot.phylo(tree,label.offset=0.1,tip.col=col, type=type.dendro)
762 for (i in rev(tree.order)) {
763 par(mar=c(0,0,1,0),cex=0.9)
764 wordcloud(names(lclasses[[i]]), lclasses[[i]], scale = c(2.5, 0.5), random.order=FALSE, colors = colcloud[i])
768 plot.dendropr <- function(tree, classes, type.dendro="phylogram", histo=FALSE, from.cmd=FALSE, bw=FALSE, lab = NULL, tclasse=TRUE) {
769 classes<-classes[classes!=0]
770 classes<-as.factor(classes)
771 sum.cl<-as.matrix(summary(classes, maxsum=1000000))
772 sum.cl<-(sum.cl/colSums(sum.cl)*100)
773 sum.cl<-round(sum.cl,2)
774 sum.cl<-cbind(sum.cl,as.matrix(100-sum.cl[,1]))
775 tree.order<- as.numeric(tree$tip.label)
779 col <- rainbow(nrow(sum.cl))[as.numeric(tree$tip.label)]
780 col <- del.yellow(col)
782 col.pie <- rainbow(nrow(sum.cl))
783 col.pie <- del.yellow(col.pie)
784 #col.vec<-rainbow(nrow(sum.cl))[as.numeric(tree[[2]])]
788 col.pie <- rep('grey',nrow(sum.cl))
791 for (i in 1:nrow(sum.cl)) vec.mat<-append(vec.mat,1)
793 for (i in 1:nrow(sum.cl)) {
794 vec.mat<-append(vec.mat,v)
800 layout(matrix(vec.mat,nrow(sum.cl),2),widths=c(3,1))
802 layout(matrix(c(1,2),1,byrow=TRUE), widths=c(3,2),TRUE)
805 par(mar=c(0,0,0,0),cex=1)
806 label.ori<-tree$tip.label
808 tree$tip.label <- lab
810 tree$tip.label<-paste('classe ',tree$tip.label)
812 plot.phylo(tree,label.offset=0.1,tip.col=col, type=type.dendro)
813 #cl.order <- as.numeric(label.ori)
815 #for (i in 1:nrow(sum.cl)) {
818 for (i in rev(tree.order)) {
819 par(mar=c(0,0,1,0),cex=0.7)
820 pie(sum.cl[i,],col=c(col.pie[i],'white'),radius = 1, labels='', clockwise=TRUE, main = paste('classe ',i,' - ',sum.cl[i,1],'%' ))
825 to.plot <- sum.cl[tree.order,1]
826 d <- barplot(to.plot,horiz=TRUE, col=col.bars, names.arg='', axes=FALSE, axisname=FALSE)
827 text(x=to.plot, y=d[,1], label=paste(round(to.plot,1),'%'), adj=1.2)
830 if (!from.cmd) dev.off()
833 #tree <- tree.cut1$tree.cl
835 plot.dendro.lex <- function(tree, to.plot, bw=FALSE, lab=NULL, lay.width=c(3,3,2), colbar=NULL, classes=NULL, direction = 'rightwards', cmd=FALSE) {
836 tree.order<- as.numeric(tree$tip.label)
837 if (!is.null(classes)) {
838 classes<-classes[classes!=0]
839 classes<-as.factor(classes)
840 sum.cl<-as.matrix(summary(classes, maxsum=1000000))
841 sum.cl<-(sum.cl/colSums(sum.cl)*100)
842 sum.cl<-round(sum.cl,2)
843 sum.cl<-cbind(sum.cl,as.matrix(100-sum.cl[,1]))
846 if (direction == 'rightwards') {
850 if (!is.null(classes)) {
851 matlay <- matrix(c(1,2,3,4),1,byrow=TRUE)
852 lay.width <- c(3,2,3,2)
854 matlay <- matrix(c(1,2,3),1,byrow=TRUE)
860 if (!is.null(classes)) {
861 matlay <- matrix(c(1,2,3,4,4,4),3)
863 matlay <- matrix(c(1,2,3,3),2)
867 layout(matlay, widths=lay.width,TRUE)
868 par(mar=c(3,0,2,4),cex=1)
869 label.ori<-tree$tip.label
871 tree$tip.label <- lab
873 tree$tip.label<-paste('classe ',tree$tip.label)
875 to.plot <- matrix(to.plot[,tree.order], nrow=nrow(to.plot), dimnames=list(rownames(to.plot), colnames(to.plot)))
877 col <- rainbow(ncol(to.plot))
878 col <- del.yellow(col)
879 if (is.null(colbar)) {
880 col.bars <- rainbow(nrow(to.plot))
881 col.bars <- del.yellow(col.bars)
887 col.bars <- grey.colors(nrow(to.plot),0,0.8)
889 col <- col[tree.order]
890 plot.phylo(tree,label.offset=0.2,tip.col=col, direction = direction, srt=srt, adj = 0.5, edge.width = 2)
891 if (!is.null(classes)) {
894 to.plota <- sum.cl[tree.order,1]
895 d <- barplot(to.plota,horiz=TRUE, col=col, names.arg='', axes=FALSE, axisname=FALSE)
896 text(x=to.plota, y=d[,1], label=paste(round(to.plota,1),'%'), adj=1.2)
899 d <- barplot(to.plot,horiz=horiz, col=col.bars, beside=TRUE, names.arg='', space = c(0.1,0.6), axisname=FALSE)
904 lcoord <- apply(cc, 1, mean)
906 if (min(to.plot) < 0) {
907 amp <- abs(max(to.plot) - min(to.plot))
914 d <- signif(amp%/%10,1)
916 mn <- round(min(to.plot))
917 mx <- round(max(to.plot))
919 if ((i/d) == (i%/%d)) {
924 plot(0, axes = FALSE, pch = '')
925 legend(x = 'center' , rev(rownames(to.plot)), fill = rev(col.bars))
932 plot.spec <- function(spec, nb.word = 20) {
935 rno <- rownames(spec)
937 if (nb.word > length(rno)) {nb.word <- length(rno)}
938 for (val in 1:ncol(spec)) {
939 rn <- rno[order(spec[,val], decreasing=T)][1:nb.word]
940 score <- spec[order(spec[,val], decreasing=T),val][1:nb.word]
941 word.to.plot <- cbind(word.to.plot, rn)
942 word.size <- cbind(word.size, score)
944 mat.lay <- matrix(1:ncol(spec),nrow=1,ncol=ncol(spec))
946 for (i in 1:ncol(spec)) {
947 col <- ifelse((i %% 2) == 0, 'red', 'blue')
948 par(mar=c(0,0,1,0),cex=0.7)
950 plot(0,0,pch='', axes = FALSE)
951 vcex <- norm.vec(word.size[,i], 2, 3)
952 text(-0.9, -0.5, cn[i], cex = 1, adj=0, srt=90, col='black')
953 for (j in 1:length(word.size[,i])) {
954 yval <- yval-(strheight(word.to.plot[j,i],cex=vcex[j])+0.02)
955 text(-0.9, yval, word.to.plot[j,i], cex = vcex[j], col = col, adj=0)
962 plot.alceste.graph <- function(rdata,nd=3,layout='fruke', chilim = 2) {
964 if (is.null(debsup)) {
965 tab.toplot<-afctable[1:(debet+1),]
966 chitab<-chistabletot[1:(debet+1),]
968 tab.toplot<-afctable[1:(debsup+1),]
969 chitab<-chistabletot[1:(debsup+1),]
971 rkeep<-select_point_chi(chitab,chilim)
972 tab.toplot<-tab.toplot[rkeep,]
973 chitab<-chitab[rkeep,]
974 dm<-dist(tab.toplot,diag=TRUE,upper=TRUE)
975 cn<-rownames(tab.toplot)
976 cl.toplot<-apply(chitab,1,which.max)
977 col<-rainbow(ncol(tab.toplot))[cl.toplot]
979 g1 <- graph.adjacency(as.matrix(dm), mode = 'lower', weighted = TRUE)
980 g.max<-minimum.spanning.tree(g1)
981 we<-(rowSums(tab.toplot)/max(rowSums(tab.toplot)))*2
982 #lo <- layout.fruchterman.reingold(g.max,dim=nd)
983 lo<- layout.kamada.kawai(g.max,dim=nd)
984 print(nrow(tab.toplot))
990 rglplot(g.max, vertex.label = cn, vertex.size = we*3, edge.width = 0.5, edge.color='black', vertex.label.color = col,vertex.color = col, layout = lo, vertex.label.cex = 1)
991 } else if (nd == 2) {
992 plot(g.max, vertex.label = cn, vertex.size = we, edge.width = 0.5, edge.color='black', vertex.label.color = col,vertex.color = col, layout = lo, vertex.label.cex = 0.8)
997 make.simi.afc <- function(x,chitable,lim=0, alpha = 0.1, movie = NULL) {
999 chimax<-as.matrix(apply(chitable,1,max))
1000 chimax<-as.matrix(chimax[,1][1:nrow(x)])
1001 chimax<-cbind(chimax,1:nrow(x))
1002 order_chi<-as.matrix(chimax[order(chimax[,1],decreasing = TRUE),])
1003 if ((lim == 0) || (lim>nrow(x))) lim <- nrow(x)
1004 x<-x[order_chi[,2][1:lim],]
1005 maxchi <- chimax[order_chi[,2][1:lim],1]
1006 #-------------------------------------------------------
1008 distm<-dist(x,diag=TRUE)
1009 distm<-as.matrix(distm)
1010 g1<-graph.adjacency(distm,mode='lower',weighted=TRUE)
1011 g1<-minimum.spanning.tree(g1)
1012 lo<-layout.kamada.kawai(g1,dim=3)
1013 lo <- layout.norm(lo, -3, 3, -3, 3, -3, 3)
1014 mc<-rainbow(ncol(chistabletot))
1015 chitable<-chitable[order_chi[,2][1:lim],]
1016 cc <- apply(chitable, 1, which.max)
1018 #mass<-(rowSums(x)/max(rowSums(x))) * 5
1019 maxchi<-norm.vec(maxchi, 0.03, 0.3)
1020 rglplot(g1,vertex.label = vire.nonascii(rownames(x)),vertex.label.color= cc,vertex.label.cex = maxchi, vertex.size = 0.1, layout=lo, rescale=FALSE)
1021 text3d(lo[,1], lo[,2],lo[,3], rownames(x), cex=maxchi, col=cc)
1022 #rgl.spheres(lo, col = cc, radius = maxchi, alpha = alpha)
1023 rgl.bg(color = c('white','black'))
1024 if (!is.null(movie)) {
1026 ReturnVal <- tkmessageBox(title="RGL 3 D",message="Cliquez pour commencer le film",icon="info",type="ok")
1028 movie3d(spin3d(axis=c(0,1,0),rpm=6), movie = 'film_graph', frames = "tmpfilm", duration=10, clean=TRUE, top = TRUE, dir = movie)
1029 ReturnVal <- tkmessageBox(title="RGL 3 D",message="Film fini !",icon="info",type="ok")
1031 while (rgl.cur() != 0)
1037 norm.vec <- function(v, min, max) {
1043 fac <- (max-min)/(vr[2]-vr[1])
1045 (v-vr[1]) * fac + min
1049 vire.nonascii <- function(rnames) {
1050 print('vire non ascii')
1051 couple <- list(c('é','e'),
1073 rnames<-gsub(c[1],c[2], rnames)
1080 #par(mar=c(0,0,0,0))
1081 #layout(matrix(c(1,2),1,byrow=TRUE), widths=c(3,2),TRUE)
1082 #par(mar=c(1,0,1,0), cex=1)
1083 #plot.phylo(tree,label.offset=0.1)
1084 #par(mar=c(0,0,0,1))
1085 #to.plot <- sum.cl[cl.order,1]
1086 #d <- barplot(to.plot,horiz=TRUE, names.arg='', axes=FALSE, axisname=FALSE)
1087 #text(x=to.plot, y=d[,1], label=round(to.plot,1), adj=1.2)
1089 make.afc.attributes <- function(rn, afc.table, contafc, clnb, column = FALSE, x=1, y=2) {
1092 afc.res <- afc.table$ligne
1093 #tokeep <- which(row.names(afc.res) %in% rn)
1094 afc.res <- afc.res[rn,]
1095 debcor <- (nd*2) + 1
1096 cor <- afc.res[,debcor:(debcor+nd-1)][,c(x,y)]
1097 debctr <- (nd*3) + 1
1098 ctr <- afc.res[,debctr:(debctr+nd-1)][,c(x,y)]
1099 massdeb <- (nd*4) + 1
1100 mass <- afc.res[,massdeb]
1101 chideb <- massdeb + 1
1102 chi <- afc.res[,chideb]
1103 inertiadeb <- chideb + 1
1104 inertia <- afc.res[,inertiadeb]
1105 frequence <- rowSums(contafc[rn,])
1107 res <- list(frequence=frequence, cor, ctr, mass = mass, chi=chi, inertia=inertia)
1112 afctogexf <- function(fileout, toplot, classes, clnb, sizes, nodes.attr=NULL) {
1113 toplot <- toplot[,1:3]
1115 #toplot <- afc$rowcoord[1:100,1:3]
1117 #rownames(toplot)<-afc$rownames[1:100]
1118 cc <- rainbow(clnb)[classes]
1119 cc <- t(sapply(cc, col2rgb, alpha=TRUE))
1120 #sizes <- apply(chistabletot[1:100,], 1, max)
1122 nodes <- data.frame(cbind(1:nrow(toplot), rownames(toplot)))
1123 colnames(nodes) <- c('id', 'label')
1124 nodes[,1] <- as.character(nodes[,1])
1125 nodes[,2] <- as.character(nodes[,2])
1127 if (! is.null(nodes.attr)) {
1128 nodesatt <- as.data.frame(nodes.attr)
1130 nodesatt <- data.frame(cbind(toplot[,1],toplot[,2]))
1133 edges<-matrix(c(1,1),ncol=2)
1134 xmin <- min(toplot[,1])
1135 xmax <- max(toplot[,1])
1136 ymin <- min(toplot[,2])
1137 ymax <- max(toplot[,2])
1138 nodes<-rbind(nodes, c(nrow(nodes)+1, 'F1'))
1139 nodes<-rbind(nodes, c(nrow(nodes)+1, 'F1'))
1140 nodes<-rbind(nodes, c(nrow(nodes)+1, 'F2'))
1141 nodes<-rbind(nodes, c(nrow(nodes)+1, 'F2'))
1142 nodesatt<-rbind(nodesatt, c(0,0))
1143 nodesatt<-rbind(nodesatt, c(0,0))
1144 nodesatt<-rbind(nodesatt, c(0,0))
1145 nodesatt<-rbind(nodesatt, c(0,0))
1146 toplot <- rbind(toplot, c(xmin, 0,0))
1147 toplot <- rbind(toplot, c(xmax,0,0))
1148 toplot <- rbind(toplot, c(0,ymin,0))
1149 toplot <- rbind(toplot, c(0,ymax,0))
1150 cc <- rbind(cc, c(255,255,255,1))
1151 cc <- rbind(cc, c(255,255,255,1))
1152 cc <- rbind(cc, c(255,255,255,1))
1153 cc <- rbind(cc, c(255,255,255,1))
1154 sizes <- c(sizes, c(0.5, 0.5, 0.5, 0.5))
1155 edges <- rbind(edges, c(nrow(nodes)-3, nrow(nodes)-2))
1156 edges <- rbind(edges, c(nrow(nodes)-1, nrow(nodes)))
1157 write.gexf(nodes, edges, output=fileout, nodesAtt=nodesatt, nodesVizAtt=list(color=cc, position=toplot, size=sizes))
1160 simi.to.gexf <- function(fileout, graph.simi, nodes.attr = NULL) {
1161 lo <- graph.simi$layout
1162 if (ncol(lo) == 3) {
1165 lo <- cbind(lo, rep(0,nrow(lo)))
1167 g <- graph.simi$graph
1168 nodes <- data.frame(cbind(1:nrow(lo), V(g)$name))
1169 colnames(nodes) <- c('id', 'label')
1170 if (! is.null(nodes.attr)) {
1171 nodesatt <- as.data.frame(nodes.attr)
1173 nodesatt <- data.frame(cbind(lo[,1],lo[,2]))
1175 edges <- as.data.frame(get.edges(g, c(1:ecount(g))))
1176 col <- graph.simi$color
1177 col <- t(sapply(col, col2rgb, alpha=TRUE))
1178 write.gexf(nodes, edges, output=fileout, nodesAtt=nodesatt, nodesVizAtt=list(color=col,position=lo, size=graph.simi$label.cex), edgesVizAtt=list(size=graph.simi$we.width))
1181 graphml.to.file <- function(graph.path) {
1183 g <- read.graph(graph.path, format='graphml')
1184 layout <- layout.fruchterman.reingold(g, dim=3)
1186 graph.simi <- list(graph=g, layout=layout, color = V(g)$color ,eff=V(g)$weight)
1192 graph.to.file <- function(graph.simi, nodesfile = NULL, edgesfile = NULL, community = FALSE, color = NULL, sweight = NULL) {
1194 g <- graph.simi$graph
1195 print(graph.simi$eff)
1196 if (!is.null(graph.simi$eff)) {
1197 V(g)$weight <- graph.simi$eff
1199 V(g)$weight <- graph.simi$label.cex
1201 layout <- layout.norm(graph.simi$layout,-5,5,-5,5,-5,5)
1203 V(g)$x <- layout[,1]
1204 V(g)$y <- layout[,2]
1205 if (ncol(layout) == 3) {
1206 V(g)$z <- layout[,3]
1209 member <- graph.simi$communities$membership
1210 col <- rainbow(max(member))
1211 v.colors <- col[member]
1212 v.colors <- col2rgb(v.colors)
1213 V(g)$r <- v.colors[1,]
1214 V(g)$g <- v.colors[2,]
1215 V(g)$b <- v.colors[3,]
1217 if (!is.null(color)) {
1218 v.colors <- col2rgb(color)
1219 V(g)$r <- v.colors[1,]
1220 V(g)$g <- v.colors[2,]
1221 V(g)$b <- v.colors[3,]
1223 if (!is.null(sweight)) {
1224 V(g)$sweight <- sweight
1226 df <- get.data.frame(g, what='both')
1227 if (!is.null(nodesfile)) {
1228 write.table(df$vertices, nodesfile, sep='\t', row.names=FALSE)
1230 if (!is.null(edgesfile)) {
1231 write.table(df$edges, edgesfile, sep='\t', row.names=FALSE)
1233 if (is.null(edgesfile) & is.null(nodesfile)) {
1238 graph.to.file2 <- function(graph, layout, nodesfile = NULL, edgesfile = NULL, community = FALSE, color = NULL, sweight = NULL) {
1241 V(g)$x <- layout[,1]
1242 V(g)$y <- layout[,2]
1243 if (ncol(layout) == 3) {
1244 V(g)$z <- layout[,3]
1246 v.colors <- col2rgb(V(g)$color)
1247 V(g)$r <- v.colors[1,]
1248 V(g)$g <- v.colors[2,]
1249 V(g)$b <- v.colors[3,]
1251 if (!is.null(sweight)) {
1252 V(g)$sweight <- sweight
1254 if (is.null(V(g)$weight)) {
1255 if (!is.null(sweight)) {
1256 V(g)$weight <- sweight
1261 df <- get.data.frame(g, what='both')
1262 if (!is.null(nodesfile)) {
1263 write.table(df$vertices, nodesfile, sep='\t', row.names=FALSE)
1265 if (!is.null(edgesfile)) {
1266 write.table(df$edges, edgesfile, sep='\t', row.names=FALSE)
1268 if (is.null(edgesfile) & is.null(nodesfile)) {