4 Plotting chromosomes
Jupyter interactive version:
This guide shows how to plot idiograms of measured karyotypes and optionally marks.
4.2 Plot monocentrics
Initially you have to put chromosome data in data.frames.
Main three data.frames:
- One for chr. sizes (parameter
dfChrSize) - One for marks’ positions (parameter
dfMarkPos) - One (optional) for mark style (parameter
dfMarkColor)
For a way to start with only one data.frame of chr. size and marks’ pos., see Chapter 8.
data.frame of chr. size
# Example data.frame to write in R, use: (column OTU is optional if only 1 OTU)
mydfChrSize<-read.table(text=
" OTU chrName shortArmSize longArmSize
\"Species one\" 1 1.5 2.0
\"Species one\" 2 2.0 2.5
\"Species one\" 3 1.0 1.5
\"Species one\" B 2.0 3.5" , header=TRUE, stringsAsFactors=FALSE,fill=TRUE)| OTU | chrName | shortArmSize | longArmSize |
|---|---|---|---|
| Species one | 1 | 1.5 | 2.0 |
| Species one | 2 | 2.0 | 2.5 |
| Species one | 3 | 1.0 | 1.5 |
| Species one | B | 2.0 | 3.5 |
loading saved data:
If you use RStudio, in the menu “Session”, use “Set working directory” for choosing your desired folder or:
setwd("~/folder/subfolder")Open your chromosome data data.frame importing it from a .csv (read.csv) or .xls file (readxl).
mydfChrSize<-read.csv("somefile.csv")Editing a data.frame:
bigdfOfChrSize <- edit(bigdfOfChrSize, edit.row.names = FALSE)For fixing column names use:
colnames(mydfChrSize)<-c("OTU", "chrName","shortArmSize","longArmSize")Marks’ position data
Open or write your mark positions in a data.frame. This data.frame has the marks present in all karyotypes with position info. This data.frame has also the centromeric marks present in all karyotypes.
Column chrRegion defines the arm or region of occurrence (p, q, w, cen). Distance to centromere is defined in markDistCen.
# We will use column OTU if data.frame because chromosome size df has it
mydfOfMarks<-read.table(text=
" OTU chrName markName chrRegion markSize markDistCen
\"Species one\" 1 45S p NA NA # no measure (NA) means whole arm
\"Species one\" 1 5S q 0.5 0.5
\"Species one\" B \"B mark\" w NA NA # w for whole chromosome
\"Species one\" B \"cB mark\" q NA 1.0
\"Species one\" 2 cgene4 p 1 1.0
\"Species one\" 2 cgene1 q 1 1
\"Species one\" 2 cgene2 q 0.5 2.0
\"Species one\" 3 DAPI q NA 1
\"Species one\" 3 cgene3 p NA 0.5
\"Species one\" 1 DAPI pcen
\"Species one\" 3 CMA qcen
\"Species one\" 2 CMA cen", header=TRUE, stringsAsFactors=FALSE,fill=TRUE)| OTU | chrName | markName | chrRegion | markSize | markDistCen |
|---|---|---|---|---|---|
| Species one | 1 | 45S | p | NA | NA |
| Species one | 1 | 5S | q | 0.5 | 0.5 |
| Species one | B | B mark | w | NA | NA |
| Species one | B | cB mark | q | NA | 1.0 |
| Species one | 2 | cgene4 | p | 1.0 | 1.0 |
| Species one | 2 | cgene1 | q | 1.0 | 1.0 |
| Species one | 2 | cgene2 | q | 0.5 | 2.0 |
| Species one | 3 | DAPI | q | NA | 1.0 |
| Species one | 3 | cgene3 | p | NA | 0.5 |
| Species one | 1 | DAPI | pcen | NA | NA |
| Species one | 3 | CMA | qcen | NA | NA |
| Species one | 2 | CMA | cen | NA | NA |
For fixing column names use something like:
colnames(mydfMarkColor)<-c("OTU", "chrName","markName","chrRegion","markSize","markDistCen") Marks’ general data
This data.frame is optional. It has the marks present in all karyotypes without position info.
It includes color and style. If the style column is not present, param. defaultStyleMark = "square" will be used during plotting.
Note the cenStyle to make constrictions, mimicking centromeres.
# From scratch:
mydfMarkColor<-read.table(text=
" markName markColor style
5S red dots
45S chartreuse3 square
cgene1 chocolate upArrow
cgene2 salmon downArrow
cgene3 \"#056522\" cMLeft
DAPI dodgerblue cM
\"cB mark\" black cenStyle
CMA darkgoldenrod1 square
\"B mark\" black square
cgene4 cornflowerblue exProtein", header=TRUE, stringsAsFactors=FALSE,fill=TRUE)| markName | markColor | style |
|---|---|---|
| 5S | red | dots |
| 45S | chartreuse3 | square |
| cgene1 | chocolate | upArrow |
| cgene2 | salmon | downArrow |
| cgene3 | #056522 | cMLeft |
| DAPI | dodgerblue | cM |
| cB mark | black | cenStyle |
| CMA | darkgoldenrod1 | square |
| B mark | black | square |
| cgene4 | cornflowerblue | exProtein |
For fixing column names use:
colnames(mydfMarkColor)<-c("markName", "markColor","style")
# if style column is not present it will be filled with "square"Add some text to the right
We will use column note to add a note to the right of the karyotype of the specified OTU (see column OTU)
notesdf<-read.table(text=
" OTU note
\"Species one\" \"Author notes\" ", header=TRUE, stringsAsFactors=FALSE,fill=TRUE)For adding just the OTU name (from column OTU of data.frame of chr. size) to the right, use OTUasNote=TRUE
Plotting
You can plot without marks (use only 1st data.frame), but we will use all 4 data.frames created.
By default the function will calculate indices (Romero-Zarco, 1986; Watanabe et al., 1999) and morphological categories of Guerra (1986) and Levan (1964).
Use parameters chrIndex and morpho of the function plotIdiograms to modify that. See ?plotIdiograms.
The cM style of mark always adds the name as if legend="inline", even when legend="aside" (default).
# svg("mydfChrSize.svg",width=12,height=6 )
par(mar = c(0, 0, 0, 0))
plotIdiograms(dfChrSize= mydfChrSize, # chr. size data.frame
dfMarkPos= mydfOfMarks, # mark position data.frame (inc. cen.)
dfMarkColor=mydfMarkColor, # mark style d.f.
chrSpacing=.6, # separ. among chr.
distTextChr = .7, # separation among text and chr. names and ind.
orderChr = "name", # order chr. by name
karHeiSpace=2 # vertical size of karyotype including spacer
,arrowhead = .5 # proportion of head of arrow
,fixCenBorder = TRUE # use chrColor as border color of cen. or cen. marks
,legendWidth = .8 # legend item width
,legendHeight = .5 # legend item height
,markLabelSpacer = 2 # legend spacer
,lwd.mimicCen = 2.5 # constric. mark line width
,rulerPos=0 # ruler position
,ruler.tck=-0.01 # ticks of ruler size and orientation
,xPosRulerTitle = 2.5 # ruler units pos.
,markPer = "45S" # show mark % of chr. under kar.
,showMarkPos = TRUE # show position of marks under kar. See bToRemove
,bToRemove = c("B mark","cB mark") # bands to remove from pos. calc. See showMarkPos
,notes=notesdf # data.frame with notes
,notesTextSize = 1.3 # font size of notes
,notesPosX = .2 # space from chr. (right) to note
,ylimBotMod = 2 # modify ylim bottom argument
,ylimTopMod = 0 # modify ylim top argument
,xlimLeftMod = 2 # modify left xlim
,xlimRightMod = 3 # modify right xlim
,lwd.cM = 2 # thickness of cM marks
,pattern = "^c" # regex pattern to remove from mark names
,remSimiMarkLeg=TRUE # remove pseudoduplicated mark names from legend (same after pattern removal)
,legendYcoord = -1.2
# ,legend="inline" # legend next to chr.
,bannedMarkName = c("B mark","cB mark") # remove from legends
)# ; dev.off() # close .svg
Vertices when centromereSize=0 are rounded:
png("mydfChrSize2.png", width=550, height=550)
par(mar = c(0, 0, 0, 0))
plotIdiograms(dfChrSize = bigdfOfChrSize[1:8,], # chr. size data.frame
dfMarkColor = mydfMarkColor,# mark style df
dfMarkPos = bigdfOfMarks, # mark position df
centromereSize = 0, # <- HERE
squareness=3, # vertices squareness
chrSpacing = .7, # space among chr.
karHeight = 2, # karyotype rel. height
karHeiSpace=4, # vertical size of karyotype including spacer
amoSepar= 2.5, # separation among karyotype
indexIdTextSize=.8, # font size of chr. name and indices
karIndexPos = .1, # position of kar. index
markLabelSize=.7, # font size of mark legends
fixCenBorder = FALSE, # do not use chrColor as border color of cen. or cen. marks
distTextChr = .8, # separation among chr. and ind.
rulerPos= 0, # ruler position
ruler.tck=-0.01, # ticks of ruler size and orientation
xPosRulerTitle = 2.6, # ruler units pos.
xlimLeftMod = 2, # modify xlim left argument
ylimBotMod = 0.4, # modify ylim bottom argument
ylimTopMod = 0 # modify ylim top argument
,lwd.cM = 2 # thickness of cM marks
)
dev.off()
There is no need to add dfMarkColor and you can also use the parameter
mycolors (optional too), to establish marks’ colors. Colors are assigned depending on the order of marks, i.e.:
unique(mydfOfMarks$markName) Let’s use the cM style of marks. A protruding line.
cM style does not apply to centromere marks. To make something similar, use centromereSize=0, legend="inline" and fixCenBorder = FALSE.
charVectorCol <- c("tomato3","darkolivegreen4","dfsd","dodgerblue","chartreuse3")
# Modify size of kar. to use rulerInterval and ceilingFactor (>= 1.13)
quo<-9
dfOfChrSize2<-dfOfChrSize
dfOfChrSize2$shortArmSize<-dfOfChrSize$shortArmSize/quo
dfOfChrSize2$longArmSize<-dfOfChrSize$longArmSize/quo
dfOfMarks2b<-dfOfMarks2
dfOfMarks2b$markSize<-dfOfMarks2$markSize/quo
dfOfMarks2b$markDistCen<-dfOfMarks2$markDistCen/quo
png("dfOfChrSizeVector.png", width=1000, height=450)
par(mar=rep(0,4))
plotIdiograms(dfChrSize = dfOfChrSize2, # d.f. of chr. sizes
dfMarkPos = dfOfMarks2b, # d.f. of marks' positions
defaultStyleMark = "cM", # forces "cM" style in d.f dfMarkColor (exc. 5S)
mycolors = charVectorCol, # colors to use
distTextChr = 0.5, # separ. text and chr.
markLabelSize=.7, # font size for labels (legend)
lwd.cM=2, # width of cM marks
legendWidth=0.9, # legend item width
legendHeight=.5,
rulerPos= 0, # ruler position
ruler.tck=-0.01, # ruler tick orientation and length
rulerNumberSize=.5 # ruler font size
,xPosRulerTitle = 2.8 # ruler units pos.
,xlimRightMod = 1 # modify xlim right arg.
)
dev.off() # close png
4.3 Plot holocentrics
Initially you have to put your chromosome data in data.frames.
Main three data.frames:
- One for chr. sizes (parameter
dfChrSize) - One for marks’ positions (parameter
dfMarkPos) - One (optional) for mark style (parameter
dfMarkColor)
For simplicity, I call a holocen. any karyotype without centromeres.
Accordingly, they lack the columns: shortArmSize and longArmSize.
From scratch:
# Example data.frame written in R, use: (column OTU is optional if only 1 OTU)
mydfChrSizeHolo<-read.table(text=
" OTU chrName chrSize
\"Species one\" 1 6.5
\"Species one\" 2 5.0
\"Species one\" 3 4.0
\"Species one\" 4 4.0
\"Species one\" X 3.0 " , header=TRUE, stringsAsFactors=FALSE,fill=TRUE)| OTU | chrName | chrSize |
|---|---|---|
| Species one | 1 | 6.5 |
| Species one | 2 | 5.0 |
| Species one | 3 | 4.0 |
| Species one | 4 | 4.0 |
| Species one | X | 3.0 |
or loading saved data:
If you use RStudio, in menu “Session”, open “Set working directory” for choosing your desired folder or:
setwd("~/folder/subfolder")Open your chromosome data as data.frame importing it from a .csv (read.csv) or .xls file (readxl).
mydfChrSize<-read.csv("somefile.csv")For fixing column names use:
colnames(mydfChrSize)<-c("OTU", "chrName","chrSize")Get marks general data
Put your mark data in a data.frame. This data.frame has the marks present in all karyotypes without position info.
If the style column is not present, param. defaultStyleMark = "square" will be used during plotting.
# From scratch:
mydfMarkColorHolo<-read.table(text=
" markName markColor style
5S red dots
45S chartreuse3 square
gene2 salmon downArrow
gene3 \"#056522\" cMLeft
gene4 darkmagenta cM
DAPI dodgerblue square
protein cornflowerblue exProtein
B black square
gene1 chocolate upArrow
CMA darkgoldenrod1 square" , header=TRUE, stringsAsFactors=FALSE,fill=TRUE)| markName | markColor | style |
|---|---|---|
| 5S | red | dots |
| 45S | chartreuse3 | square |
| gene2 | salmon | downArrow |
| gene3 | #056522 | cMLeft |
| gene4 | darkmagenta | cM |
| DAPI | dodgerblue | square |
| protein | cornflowerblue | exProtein |
| B | black | square |
| gene1 | chocolate | upArrow |
| CMA | darkgoldenrod1 | square |
For fixing column names use:
colnames(mydfMarkColorHolo) <- c("markName", "markColor","style")
# if style column not present it will be filled with "square"Get marks positions data
Open or write your mark positions in a data.frame. This data.frame has the marks present in all karyotypes with position info.
For holocentrics, position is defined with column markPos, while in monocentrics it is markDistCen.
Column chrRegion can be absent for holocentrics.
# We will use column OTU if data.frame because chromosome size df has it
mydfMarkPosHolo<-read.table(text=
" OTU chrName markName markPos markSize chrRegion
\"Species one\" 4 B NA NA w # w = whole chromosome mark
\"Species one\" 3 DAPI 2.0 0.5
\"Species one\" 1 45S 2.0 0.5
\"Species one\" 2 protein 3.5 1.5
\"Species one\" 2 gene1 1.0 0.5
\"Species one\" 1 gene2 1.0 0.5
\"Species one\" 4 gene3 3.0 0.5
\"Species one\" 3 gene4 1.0 0.5
\"Species one\" X CMA 2.0 0.5
\"Species one\" X 5S 0.5 0.5
\"Species one\" X 5S 0.5 0.5" , header=TRUE, stringsAsFactors=FALSE,fill=TRUE)| OTU | chrName | markName | markPos | markSize | chrRegion |
|---|---|---|---|---|---|
| Species one | 4 | B | NA | NA | w |
| Species one | 3 | DAPI | 2.0 | 0.5 | |
| Species one | 1 | 45S | 2.0 | 0.5 | |
| Species one | 2 | protein | 3.5 | 1.5 | |
| Species one | 2 | gene1 | 1.0 | 0.5 | |
| Species one | 1 | gene2 | 1.0 | 0.5 | |
| Species one | 4 | gene3 | 3.0 | 0.5 | |
| Species one | 3 | gene4 | 1.0 | 0.5 | |
| Species one | X | CMA | 2.0 | 0.5 | |
| Species one | X | 5S | 0.5 | 0.5 | |
| Species one | X | 5S | 0.5 | 0.5 |
For fixing column names use something like:
colnames(mydfMarkPosHolo)<-c("OTU", "chrName","markName","markPos","markSize") Plotting
You can plot without marks (passing only 1st data.frame), but we will use all 4 data.frames created:
# library(idiogramFISH)
# svg("mydfChrSizeHolo.svg",width=13.5,height=6 )
# png("mydChrSizeHolo.png", width=600, height=300)
par(mar=c(0,0,0,1)) # bottom left top right
plotIdiograms(dfChrSize = mydfChrSizeHolo, # data.frame of chr. sizes
dfMarkColor= mydfMarkColorHolo, # df of mark style
dfMarkPos = mydfMarkPosHolo, # df of mark positions
chrSpacing = 1,
addOTUName=FALSE, # add OTU names
xlimLeftMod= 2, # modify xlim left argument
ylimTopMod= -1, # modify ylim top argument
ylimBotMod= -2 # modify ylim bottom argument
,rulerPos = 0 # ruler position
,ruler.tck = -0.01 # ruler ticks size and orient.
,xPosRulerTitle=2.6
,legendWidth=1 # width of legend
,legendHeight=.5 # height of legend item
,xModifier = 50 # separation among chromatids
); # dev.off() # closes png or svg
It is not mandatory to use dfMarkColor and you can also use the parameter
mycolors (optional too), to establish marks’ colors. When using mycolors, colors are assigned depending on the order of marks, i.e.:
unique(dfMarkPosHolo$markName)
# [1] "5S" "DAPI" "45S" "CMA"
par(mar=rep(0,4))
plotIdiograms(dfChrSize = dfChrSizeHolo, # d.f. of chr. size
dfMarkPos = dfMarkPosHolo, # d.f. of marks' positions
mycolors = c("red","dodgerblue","fdsjkfds","chartreuse3","darkgoldenrod1"), # colors for marks
addOTUName=FALSE, # do not add OTU name
ruler=FALSE, # do not add ruler
xlimLeftMod=1, # modify left xlim arg.
xlimRightMod=3, # modify right xlim arg.
ylimBotMod=.2 # modify bottom ylim
,chromatids=FALSE # do not show separ. chromatids
)
4.4 ggplot holocentrics
With function mapGGChrMark and the data.frames of chr. (dfChrSizeHolo) and marks’ position (dfMarkPosHolo) for one OTU, you get a list of data.frames for ggplot (Wickham et al., 2020)
chrAndMarksMap <- mapGGChrMark(dfChrSizeHolo, dfMarkPosHolo, chrSpacing = .5)
require(ggplot2)
myColors <- c("chartreuse3","red","darkgoldenrod1","dodgerblue")
names(myColors) <- c("45S","5S","CMA","DAPI")
ggplot() +
geom_polygon(aes(x=x,
y=y,
group=Chr
)
,data=chrAndMarksMap$dataChr
,color="gray"
,fill="gray"
) +
geom_polygon(aes(x=x,y=y,
group=id,
color=markName,
fill=markName
)
,data=chrAndMarksMap$dataMark
) +
theme_classic() +
scale_color_manual(
values = myColors
) +
scale_fill_manual(
values = myColors
) +
scale_x_continuous(breaks = seq(1, nrow(dfChrSizeHolo), 1 )
) +
scale_y_continuous(breaks = seq(0, 5, 0.5)
) +
geom_segment(aes(y=0, yend=5, x=-Inf, xend=-Inf)
)+
theme(axis.line = element_blank(),
axis.ticks.x= element_blank(),
axis.title = element_blank()
,title = element_blank()
)
require(svglite)
ggsave(file="ggplot.svg",width=6, height=4)
4.5 Plot holo. and mono. in the same karyotype
To accomplish this, we will use cenStyle marks in a “holocen.” karyotype. The ruler is continuous as in holocen.
# mark general characteristics' data.frame:
mydfMarkColorHolo2 <- read.table(text=
" markName markColor style
5S red dots
45S chartreuse3 square
gene2 salmon downArrow
gene3 \"#056522\" cMLeft
gene4 darkmagenta cM
DAPI dodgerblue square
protein cornflowerblue exProtein
B black square
gene1 chocolate upArrow
CMA darkgoldenrod1 square
c45S chartreuse3 cenStyle # <- simulate Cen.
myCen2 red cenStyle
myCen chartreuse2 cenStyle
constr NA cenStyle" , header=TRUE, stringsAsFactors=FALSE,fill=TRUE)
# add new cenStyle marks to data.frame of marks' position, created above
mydfMarkPosHolo2<-plyr::rbind.fill(mydfMarkPosHolo,
data.frame(OTU="Species one",
chrName=c(1:3,"X"),
markName=c("c45S","myCen2","constr","myCen"), # <- use new mark
markPos= c(rep(2.5,3),1),
markSize=NA
)
)
# png("mydChrSizeHolo.png", width=600, height=300)
par(mar=c(0,0,0,1)) # bottom left top right
plotIdiograms(dfChrSize = mydfChrSizeHolo, # data.frame of chr. sizes
dfMarkColor= mydfMarkColorHolo2, # df of mark style
dfMarkPos = mydfMarkPosHolo2, # df of mark positions
addOTUName=FALSE, # add OTU names
xlimLeftMod= 2, # modify xlim left argument
ylimTopMod= -1, # modify ylim top argument
ylimBotMod= -2 # modify ylim bottom argument
,rulerPos = 0
,ruler.tck = -0.01
,xPosRulerTitle = 2.6
,legendWidth=1 # width of legend
,legendHeight=.5 # height of legend item
,lwd.mimicCen=2.5 # line width of const. mark
,pattern="^c" # regex pattern to remove from mark names
,remSimiMarkLeg = TRUE # remove pseudoduplicated mark names (got equal after pattern removal)
,bannedMarkName = c("myCen","myCen2","B") # hide label
# ,legend="inline" # legends inline
) #; dev.off() closes png 