There are 4 helper functions in scales used to demonstrate ggplot2 style scales for specific types of data: demo_continuous () and demo_log10 () for numerical axes. scale_y_continuous é usado para definir valores para a estética da escala do eixo y contínuo. The function scale_y_continuous allows for functions to be used for the labels argument. Every plot has two position scales, corresponding to the x and y aesthetics. scale_x_continuous(), scale_y_continuous()의 이해와 표현 ggplot() 함수와 함께 사용할 수 있는 scale_x_continuous(), scale_y_continuous() 함수는 연속하는 숫자형 변수 x,y에 대하여 각각 축의 스케일(scale),. r; ggplot2; contour; Share. 14. Unlike most {ggplot2} functions, scales are not additive. this modified code should work. This means that if a scale attribute is modified in one call to scale_x_continuous(), a second call to scale_x_continuous() will write over all changes made in the first. 5. Thus, using percent() is not an option anymore. The difference between the two (irrelevant for geom_point) is that scale_y_continuous (limits = c (. If you need to include the whiskers as well, consider using boxplot. A menudo, es posible que desee convertir el eje x o la escala del eje y de un gráfico ggplot2 en una escala logarítmica. Sorted by: 39. Additionally, you can't use _scale_continuous for a factor. Trying to format y axis to display thousands to 1 decimal place. 1,1), expand = c (0,0)) and I get this as the result. Here is the issue: I am using ggplot to to make a graph in which the x axis ranges from -90 to 90 degrees. Every continuous scale takes a trans argument, allowing the use of a variety of transformations: The transformation. e. 1. Mar 18, 2022 at 14:05. library (ggplot2) library (lemon) ggplot (diamonds, aes (x=price, fill=cut)) + geom_histogram. I had previously been doing this using: scale_x_discrete (labels=c ("old_label" = "new_label")) However, I cannot use both scale_x_discrete. percent_format() and percent() multiply values by one hundred and display percent sign. Just do fivenum() on the data to extract what, IIRC, is used for the upper and lower hinges on boxplots and use that output in the scale_y_continuous() call that @Ritchie showed. See how to set custom axis breaks, number of breaks, labels and limits for different scenarios. library (dplyr) library (ggplot2) mtcars %>% count (cyl) %>% mutate (prop = n / sum (n)) %>% ggplot (aes (x = cyl, y = prop)) + geom_point () + scale_y_continuous. scale_x_continuous and scale_y_continuous are the default scales for continuous x and y aesthetics. R ggplot2 scale_y_continuous : Combining breaks & limits. This will extend only the right end of your Y-axis by 10% (. Afterwards you could get petrcent labels using scales::percent:Complete noob to R/RStudio/tidyverse. Breaks in scale_x_continuous doesn't seem to work. percent_format() and percent() multiply values by one hundred and display percent sign. To make both changes work, get rid of ylim () and set both limits and breaks in scale_y_continuous (): pg_plot + scale_y_continuous(limits = c(0, 10), breaks =. I tried using scale_y_continuous but it erased the entire y-labels. scale_y_continuous() followed by scale_y_reverse(), the first scale is overridden. 05, 0)", instead of "c(0, 0)". + 10)) # Inherit the name from the primary. Goal: change labels on my y axis on a bar plot from e. library (ggplot2) ggplot () + geom_col ( data = f400weight, aes (factor (month), avg_weight, fill = factor (fruit_origin. let me look. Continuous y position for datetime data points. 0 "You Stupid Darkness" / RStudio 1. y. Francesco1 August 13, 2021, 1:27pm #1. Set up data: set. 1 Continuous Axis. A convenient way to specify what guides should be drawn where is the guides. Suppose we have the following data frame in R that shows the percentage of items that were returned at four different stores:Position guides are ticks, labels and lines drawn at the x- and y-axes. scale_y_continuous( breaks=pretty_breaks(), expand = c(0. should hide the outliers. There are three ways to control the plot limits: Adjusting what data are plotted. I want to do this inside a ggplot: scale_y_continuous (labels = function (l) { trans = l / 1000, paste0 (l, "K") }) If I add either of the two commands alone, it works, i. I'm using the geom_smooth function for the regression line, but I need 2 regression lines (one for each species). Improve this answer. Lin. It just goes against the math definition. NOTE it's important to add 0 to the breaks to make it. You could make a special case for 1e-5 using prettyNum0 <. I have successfully used a function to add degree symbols to the tick labels created by scale_x_continuous. 2. Looking at log_trans and reverse_trans in the scales package for guidance and inspiration, a. It's because you are setting a discrete x scale but your x values are numeric. This is a shortcut for supplying the limits argument to the individual scales. demo_datetime for data / time axes. scale_y_continuous를 사용하여 R의 백분율로 Y 축 레이블 인쇄. breaks, labels, limits,. #> Warning: Removed 25 rows containing. This is useful for scales which span. get_breaks. 2. This is always scales::rescale (), except for diverging and n colour gradients (i. 5. limit,upper. You can fix the ends of the color bar by giving a limits argument to the scale; it should cover the whole range that the data can. 3. 1 Answer. When asking a second question: Instead of pasting the answer in your post simply put a link to the first question/answer in the post. 8, "1. ggplot(dt,aes(x=XVal,y=YVal)) + geom_line(aes(color=Type)) + facet_wrap(~Grp,scales = "free_y", ncol = 2) + scale_y_continuous(breaks = my_breaks, labels = function(x){round(x,2)}) Notice, however that in Group C, the labels end up not making total sense, since both values for the breaks (0. Variable data is continuous data, this means that the data values can be any real number like 2. 6. Your bars starts at 0 point and therefore are removed because minimal y value is set higher. Continuous Data. I have a peculiar problem involving the use of the trans and label commands in scale_y_continuous within ggplot2. Setting the limits in each scale. library(tidyverse) df <- mpg %>% head() %>% mutate(hwy = hwy * 10000) ggplot(df, aes(cty, hwy)) + geom_point() + scale_y_continuous(label = scales::comma) + geom_text. , scale_x_continuous(trans = "log10"). The default replaces out of bounds values with NA. # Show colorbar guide for colour. Then the limits get set to c(0,0. lab = and then reformatting the y axis labels with str_replace_all from stringr. e. vector of multiplicative range expansion factors. 5. By default, the tick labels will be the same as the breaks, but you can change that with the labels argument, either by giving the labels you want as a character string or by giving a function to run on the breaks values. However, when I adjust the font face using theme (), the tick labels do not become bold. Note: In the examples below, where it says something like scale_y_continuous, scale_x_continuous, or ylim, the y can be replaced with x if you want to operate on the other axis. scales::percent(100, scale = 1) ## [1] "100%" 然而，scale_y_continuous()中的labels参数期望的是一个函数而非一个实际的标签值作为其输入，引起使用percent()不是一个好的选项。不过好在scales包也提供了另一个percent_format()函数，它可以返回一个已经更改过默认值的percent()函数。Some common formats are built into the scales package: x <- rnorm (10) * 100000 y <- seq (0, 1, length = 10) p <- qplot (x, y) library (scales) p + scale_y_continuous (labels = percent) p + scale_y_continuous (labels = dollar) p + scale_x_continuous (labels = comma) # qplot allows you to do some of this with a little less typing: # * axis. Note that these facets must be used with scales = "free" or "free_x" or "free_y", depending on what scales are added. 1 Making a Basic Line Graph. e. + scale_y_continuous(labels = scales::percent) However I have not been able to find how to do this in Plotnine. a grid::unit() object specifying the length of the short tick marks. 使用的函数是 scale_y_continuous( ) ，它是ggplot2库中 “y-aesthetics “的一个默认比例。由于我们需要在Y轴的标签中加入百分比，所以使用了关键词 “labels “。 现在使用 scales: : percent 将Y轴的标签转换成百分比。这将把Y轴的数据从十进制扩展到百分比。I have 40 groups (defined by short_ID) and would like to produce 40 different plots that use different y-scale breaks for each short_ID. I'm using : scale_y_continuous(labels = scales::unit_format("k", 1e-3)) but displays as a whole number. Here's an explanation: First, The breaks argument in scale_y_continuous() can take the form of a function of the plot's input data (x in this case) Second, seq(0, (max(x) + 1) * 1. Value. In the example below the transformation for the secondary axis. Visualization examples • povcalnetR - GitHub Pages. Only a logarithmic function has the property that f(10^-5) - f(10^-4) == f(10^-4) - f(10^-3). To remove this gap currently one has to add scale_y_continuous(expand = expansion(c(0, 0. 0. 12, 3. 1))trans="log10" and labels = scales::dollar problem. You can try: # To deactivate scientific notation on y-axis: p + scale_y_continuous(labels = function(x) format(x, scientific = FALSE)) # To activate scientific notation on y-axis: p + scale_y_continuous(labels = function(x) format(x, scientific = TRUE)) # To deactivate. However you can create a pseudolog scale using scales::pseudo_log_trans to get 0 included on the axis so all the bars go the same direciton. This can be done in a number of ways, as described on this page. Below I've illustrated how this can be done using the mtcars dataset. Powered by. Position scales for continuous data (x & y) Description. Length)) + geom_histogram() + scale_y_continuous(expand = c(0. This is a convenience function for generating scale expansion vectors for the expand argument of scale_ (x|y)_continuous and scale_ (x|y)_discrete. 4) for 40%:Method 1: Whole number representation. Learn how to use the scale_y_continuous function in R to set values, print labels, modify scaling ratio, remove labels or customize labels for continuous y-axis scale. Follow edited Oct 8, 2018 at 3:23. Also accepts rlang lambda function notation. The expansion vectors are used to add some space between the data and the axes. 3)) p A function that takes the breaks as input and returns labels as output. There are three variants that set the trans argument for commonly used transformations: scale_*_log10(), scale_*_sqrt() and scale_*_reverse(). 0. In this R tutorial you’ll learn how to set the axis labels of a barchart in percentage points. Disclaimer: I'm the author of ggh4x. However, sometimes it may be preferable to adjust only one side of an axis and keep the default value for the other side. ie, since the y-axis is transformed using ~. Use scale_y_continuous para remover rótulos no eixo Y em R. scale_y_continuous 는 연속적인 y 축 스케일 미학을위한 값을 설정하는 데 사용됩니다. In the scale_y_continuou () function there is an argument sec. Basics. ). Hi there, I need some help. 3. A date-time value will create a continuous date/time scale. In most. Additional text to display before the number. scale_y_discrete A handy way to supply some sample data is the dput() function. ehl November 3, 2022, 3:24pm #1. The issue is that you are plotting the counts. e. + scale_y_continuous(labels = scales::percent) Or, to specify formatting parameters for the percent: + scale_y_continuous(labels = scales::percent_format(accuracy = 1)) (the command labels = percent is obsolete since version 2. right = element_line (color = "red")) Maybe there is a. Here are 2 graphs that I made using a small sample of my datagraphs. This is cumbersome to type, easy to forget and hard to grasp for beginners. All label_ () functions return a "labelling" function, i. continuous_scale: Continuous scale constructor; coord_cartesian: Cartesian coordinatesThe "error" you discussed is actually just a warning, because you used both ylim and then scale_y_continuous. 3, scale_y_continuous (expand = expansion (mult = c (0, . 4 Answers. 6) for discrete variables. 이 예에서는 scale_y_continuous 를 사용하여 Y 축 레이블을. This answer is out of date for ggplot2 version 0. scale_y_continuous (breaks = seq (-100, 100, 2), labels = abs (seq (-100, 100, 2))). 90. First, this simple code should yield the following figure: ggplot (data = mpg, aes (x = displ, y = hwy)) + geom_point () + scale_x_continuous (sec. From experience, I wrote how I’d shown a chart over many years of the regional mortality with the left axis and the Trust mortality numbers on. In ggplot2 version 3. Learn how to customize the y-axis of a plot using the scale_y_continuous function in ggplot2 with examples and syntax. 3. A numeric value will create a continuous scale. + coord_sf(expand = F) + scale_x_continuous(breaks = c(33. Control of the x and y axes for continuous variables is done with the functions scale_x_continuous and scale_y_continuous. This started happening a few days ago, that scales::percent would add a decimal place in its labels, and I can't seem to disable this decimal to display integer values on y-axis. I hope this helps understanding why this plot is giving you trouble. Description. comma_format() and comma() format numbers with commas separating thousands. I'm an absolute beginer in ggplot but I need to rapidly expand the size of the y-labels in a code written by a colleague which is currently unreachable. This article tells us that “it’s OK not to start your y-axis at zero”, but then states that “column and bar charts should always have zeroed axes”. Using R 3. A question was posed on the NHS-R Slack asking for help to code 2 scale y axes on a {ggplot2} chart. 0. First. timedelta64 (1, 's') The graph can properly scaled with:11. Now I try to transform the Y axis according to the distribution used. Scale transformation. This answer is out of date for ggplot2 version 0. I see. I want to make a ggplot for which the y-axis labels are formatted by a pre-made list. With scales you can make use of trans_new to define a new transformation. Question: how to utilize n or similar to create two text-lines in x-axis label written with paste0 I have . comes up with error: Error: Discrete value supplied to continuous scale. In the following. See the arguments, examples and built-in transformations for each variant. This works perfectly! But now, I want to categorise the items in the histogram, as follows: ggplot (contig_len, aes (x = Length, fill = Prevalence)) + geom_histogram (binwidth=200, alpha=0. scale_y_continuous (limits=c (-5, 1)) # or whatever values you want to use. , scale_colour_gradient2 () ,. My goal is to round to either 1 decimal (51. I start with theme_classic() then make modifications using theme(). 이 함수는ggplot2 패키지의 일부이며 대부분ggplot 객체와 함께 사용되어 그릴 그래프에 대해 다른 매개 변수를 수정합니다. na. Break points not behaving with scale_y_continuous() 0. @konvas If I use scale_y_continuous (breaks = c (0, 1, 3. In the simplest case they map linearly from the data. Colour gradients are often used to show the height of a 2d surface. This follows for all other places you define those limits. 5 Coloring Negative and Positive Bars Differently. Based on these functions trans_new is defined. As of v3. Cómo crear una escala logarítmica en ggplot2. 6 units on each side for discrete variables. #' `scale_x_binned ()` and `scale_y_binned ()` are scales that discretize. the log2-transformed fold change. The idea is to increase at least +1 to the maximum value of the plot with the highest y-axis value (in the case explained above, it would be the second boxplot with n=8) I have tried to change the y-axis with scale_y_continuous like this: p <- p + scale_y_continuous(limits = c(0, 5. Please mark answers as accepted if they helped you to solve your problem. How to set the coordinate limits when x and y scales are very different? 1. 1) First we make a sequence between 0 and the maximum value of the x-axis, plus some extra padding ((x+1)*1. scale_y_continuous 用于设置连续 y 轴比例美学的值。该函数是 ggplot2 包的一部分，它主要与 ggplot 对象一起使用来修改要绘制的图形的不同参数。此示例演示如何使用 scale_y_continuous 将 Y 轴标签 scale_y_continuous を使用して、R の Y 軸のスケーリング比を設定する. 1. 5,6. Labelling functions are designed to be used with the labels argument of ggplot2 scales. 2 Scale transformation. So to make sure the pretty breaks line up with the limits based on the original. This older stackoverflow question explains how to change your y-axis to K for thousands instead of ,000. ), i. You can add linetype inside aes in your geom_line call to create a separate legend for the line then move its legend closer to fill legend. x*800/50 does. I would like the numerical value yielded from seq(0,80,5) to appear in both lines, but with % written. as you can see one subset goes up to 6% and the other goes up to 2%, on my original data the Y scale goes up to 13% and 3. </p>. This occurs because the default for scale_(x|y)continuous* for continuous variables adds 5% at either end so that points are not right at the edge. # All these. Numeric columns can be reversed by adding scale_y_reverse() or scale_y_continuous(trans = "reverse) but I can't seem to figure out how to get from top to bottom: 2005, 2006, 2007. It only works with facets where scales are free. 0,0)) + ylim(0,15) Now, the histogram is no longer sitting on the x-axis. For this reason, the ggsurvfit() and ggcuminc() functions do not modify the default {ggplot2} scales; rather, all. scale_y_continuous(name="Fluorescent intensity/arbitrary units", labels = comma) to your ggplot statement. When working with continuous data, the default is to map linearly from the data space onto the aesthetic space. any data points outside the range of the limits will become NAs. 0. Use scale_y_continuous() or scale_x_continuous() ggplot(df, aes (x=x, y=y)) + geom_point() + scale_y_continuous(trans=' log10 ') + scale_x_continuous(trans='. I’ve tried several ways of introducing the “round” function into both steps 2 and steps 3 below, but I can’t get rid of these unnecessary decimals. How can I display 1 decimal place so instead of 30k, I get 30. Setting limits on the coordinate system will zoom the plot (like you're looking at it with a magnifying glass), and will not change the. Follow. 6 and I wnat to change it to be 0. frame (x = 1:5, y = 1:5, p = 1:5, q = factor (1:5), r = factor (1:5)) p <- ggplot (dat, aes (x, y, colour = p, size = q, shape = r)) + geom_point () # without guide specification p #> Warning: Using size for a discrete variable is not advised. , scale_x_continuous(trans = "log10"). , without needing to change the the original function to output log10 values). Next, we will create a function using a series of if else statements to “gradually” identify the individual facet panels based on their current limits, and then set the new limits for each of them. The scale_x_continuous () and scale_y_continuous () methods can be used to disable scientific notation and convert scientific labels to discrete form. ) only accepts a single scale. frame has only 2 rows. answered Jan. We recommend you read our Getting Started guide for the latest installation or upgrade instructions, then move. ))) + scale_y_continuous (formatter = 'percent') if your data has NAs and you dont want them to be. 2. A question and answers forum for R users to share and discuss their code and problems. This is useful if the underlying data is very small or very large. You should remove limits= from scale_y_continous () and use coord_cartesian () with ylim= instead. The large variation in both the Total values and the ration between Total and Failed make the graph hard to read. frame(x = 1:5, y = 1:5) p <- ggplot(df, aes(x, y)) + geom_point() p <- p + expand_limits(x = 0, y = 0) p # not what you are looking for p + scale_x_continuous(expand = c(0, 0)) + scale_y_continuous(expand = c(0, 0)) You may need to adjust things a little to make sure points are not getting cut off (see, for example, the point at x. asked Mar 6, 2014 at 15:22. Unlike continuous scales, discrete scales can easily show missing values, and do so by default. packages ("devtools") devtools::install_github ("tidyverse/ggplot2") library (ggplot2) p + theme ( axis. right after your limits =. Of course, the relative scales for the two y-axis values are different (actually should be "adjusted" according to the y values in the first dataset. . 3. super. Improve this question. Just change the first part toJust calling scale_y_continuous doesn't scale the axes, your call to breaks just specifies where on the current scale they should be set. Source: R/scale-discrete-. Sorted by: 1. 1. axis which allows you to plot a second axis on the right-hand side of the plot. This data is measured on a continual scale like distance, time, weight, length etc. Ideally, I would use the lowest value (+ some space) from both plots, and the highest value (+ some space) from both plots for the limits of both plots. Suppose we have the following data frame in R that shows the percentage of items that were returned at four different stores:ggplot (subset (mtcars, am==1), aes (x=wt, y=mpg, colour=carb)) + geom_point (size=6) In the top one, dark blue is 1 and light blue is 4, while in the bottom one, dark blue is (still) 1, but light blue is now 8. this is helpful, however, the scale from the scale_y_continuous function is applied across all boxplots. The defaults are c (0. (I know this is somewhat abstract; see the below code to get a better. library (ggplot2) library (scales) nminor <- 7 nmajor <- 5 ggplot (iris, aes (x = Species, y = Sepal. 1. This is the basic boxplot that we will work with, using the built-in PlantGrowth data set. short. R R Plot. 1 The “tidy” approach to data visualization. . 3. ) and as a function labels = percent. The one problem with this solution is that the. 5. ggplot2, rstudio. Minor suggested edit to the response above: It seems that you have to specify the limits within the scale_y_continuous call prior to setting the values as percentages: scale_y_continuous (limits=c (0,1), labels = scales::percent) Share. Jblum Jblum. a grid::unit() object specifying the length of the middle tick. The exponential constant e is positive, and y is just an exponent. This behaviour depends on the oob (out-of-bounds) argument of scale_y_continuous(), which defaults to the scales::censor() function. 5), limits = c(5,7)) was the solution! What you may have to keep in mind if you use log transformed data like me is that if you simply put c(0,40) or similar, your data may appear very small as the distance from 0 to the first break (10 in my case) is large and it could be better to use the real bottom. Other position scales: scale_x_binned(), scale_x_continuous(), scale_x_date() Examples+ scale_y_continuous(labels = scales::percent) The following example show how to use this syntax in practice. ggplot2 の scale_x_continuous で x 軸の限界を設定する. To fix this problem, the expand argument within the scale_y_continuous section needs to be set to "c(0. First, how does round the number in the data label. Now, the same format would be specified (much more neatly) this way: scale_y_continuous (labels=function (x)x*1000) or if you want to use the same labelling scheme multiple times: formatter1000 <- function () { function (x)x*1000 } scale_y_continuous (labels=formatter1000. The same thing happen with the secondary y-axis, the limits for the secondary-y axis goes from 40 000 to 240 000 with 40 000(increment of sequence) instead of going from 0 to 250 000 with 50 000(increment of sequence). 5, position="stack") + scale_y_continuous (trans = "log1p") This doesn't work, however, as the stacking is performed without taking the log scale into. binned_scale: Binning scale constructor; borders: Create a layer of map borders; calc_element: Calculate the element properties, by inheriting properties. The basic steps involved are the same whichever graphics package you use: Transform the data into the Y scale that you want. The scales package, a ggplot2 dependency 4, makes it incredibly easy to reformat x and y axis labels (among other things). The dotted line would therefore look higher on the y-axis and differences between 1 and 2 would be noticeable. 4) Video & Further Resources. This needed a bit of jiggery-pokery to get the second axis on a reasonable scale. As you can see, I tried using scale_y_continuous with limits 0 and 15000 with a step of 500, but it sets the limits, and shows just the values between 0 and 500 in my graphs. # donttest { # ggplot object dat <- data. Then I try to use the same exact argument with an area plot and it screws the plot up in a. However, to reply to your question and get your scale starting at 1 instead of 0, you need to change scale_y_continuous by this: scale_y_continuous (name="Rating", breaks=1:7, limits=c (0, 7)) Does it answer your. With other kinds of plots, it seems like you can call something like scale_y_continuous(limits=c(0, 100), expand = c(0, 0)) (for example), but calling scale_linetype_manual() with these parameters. df <- data. g. Setting the limits in each scale. In most cases this is clear in the plot specification, because the user explicitly specifies the variables mapped to x and y explicitly. Note that, scale_x_continuous() and scale_y_continuous() remove all data points outside the given range and, the coord_cartesian() function only adjusts the visible area. Instead of changing the data (mutate(y = y / 10^6)), the scale argument can be used to do conversion on the fly : scale_y_continuous(labels = unit_format(unit = "M", scale = 1e-6)) – bug313. If the X and Y axis represent continuous data, we can use scale_x_continuous() and scale_y_continuous() to modify the axis. 0. breaks and 2. Here's the full code for the graph:This topic was automatically closed 21 days after the last reply. This occurs because the default for scale_(x|y)continuous* for continuous variables adds 5% at either end so that points are not right at the edge. Improve this answer. A função é parte do pacote ggplot2 e é usada principalmente com objetos ggplot para modificar diferentes parâmetros para gráficos a serem. scale_x/y_continuous breaks by n in R ggplot2? 1. ggplot2: change break points of discrete scale to be between two break points. excluding any data points outside that range), whereas coord_cartesian changes the plotting range after any calculations. Although ggplot doesn't allow creating a separate independent y-axis, it does allow creating a second y axis that is a one-to-one transformation of the first. p1 <- ggplot (mpg, aes (displ, hwy)) +. –. As of v3. Vanilla ggplot2 comes with two position guides: guide_axis (), which draws axes, and guide_none (), which skips drawing anything. Adding another scale for 'y', which will #> replace the existing scale. Sorted by: 2. I must have had a typo or not tried this properly somehow, but the solution is that scale_y_reverse () takes 'label' as an argument, just like scale_y_continuous. answered Dec 2, 2018 at 16:35. ternatively, you can use the function scale_y_continuous(trans = "log10"), which allows to transform breaks and the format of labels. coord_cartesian () just zooms that region of values. Below minimal examples illustrates that I can get percentage scale labels ( labels = percent ) or an absolute scale ( labels = abs ) but I have no idea how to combine them. You can move this threshold to 1e-5 with the labeller function prettyNum0 <- function(x){sprintf("%. Each aesthetic property of the graph (y-axis, x-axis, color, etc. You can also extend that end by a fixed amount: for instance, scale_y_continuous (expand = expansion (add = c (0, 5))) extends it by 5 units of space. Inspired by Stack. The same thing happen with the secondary y-axis, the limits for the secondary-y axis goes from 40 000 to 240 000 with 40 000(increment of sequence) instead of going from 0 to 250 000 with 50 000(increment of sequence). colour=NA) + coord_cartesian(ylim = c(0, 100)) From the coord_cartesian documentation:. Depending on whether one wants to modify the x or the y axis scale_x_* or scale_y_*.