{"id":851,"date":"2015-04-07T10:43:19","date_gmt":"2015-04-07T17:43:19","guid":{"rendered":"http:\/\/homepages.uc.edu\/~yaozo\/wordpress\/?p=851"},"modified":"2015-04-07T10:43:19","modified_gmt":"2015-04-07T17:43:19","slug":"how-to-format-plots-for-publication-using-ggplot2","status":"publish","type":"post","link":"https:\/\/zhuoyao.net\/index.php\/2015\/04\/07\/how-to-format-plots-for-publication-using-ggplot2\/","title":{"rendered":"How to format plots for publication using ggplot2"},"content":{"rendered":"

The following is the code from a presentation made by Rosemary Hartman to the Davis R Users\u2019 Group<\/a>. I\u2019ve run the code through the spin<\/code> function in knitr<\/code> to produce this post. Download the script to walk throughhere<\/a>.<\/em><\/strong><\/p>\n

First, make your plot. I am going to use the data already in R about sleep habits of different animals. It\u2019s the same one Noam used for his intro to ggplot.<\/a><\/p>\n

library<\/span>(ggplot2)\nstr<\/span>(msleep)<\/code><\/pre>\n
## 'data.frame':    83 obs. of  11 variables:\n##  $ name        : chr  \"Cheetah\" \"Owl monkey\" \"Mountain beaver\" \"Greater short-tailed shrew\" ...\n##  $ genus       : chr  \"Acinonyx\" \"Aotus\" \"Aplodontia\" \"Blarina\" ...\n##  $ vore        : Factor w\/ 4 levels \"carni\",\"herbi\",..: 1 4 2 4 2 2 1 NA 1 2 ...\n##  $ order       : chr  \"Carnivora\" \"Primates\" \"Rodentia\" \"Soricomorpha\" ...\n##  $ conservation: Factor w\/ 7 levels \"\",\"cd\",\"domesticated\",..: 5 NA 6 5 3 NA 7 NA 3 5 ...\n##  $ sleep_total : num  12.1 17 14.4 14.9 4 14.4 8.7 7 10.1 3 ...\n##  $ sleep_rem   : num  NA 1.8 2.4 2.3 0.7 2.2 1.4 NA 2.9 NA ...\n##  $ sleep_cycle : num  NA NA NA 0.133 0.667 ...\n##  $ awake       : num  11.9 7 9.6 9.1 20 9.6 15.3 17 13.9 21 ...\n##  $ brainwt     : num  NA 0.0155 NA 0.00029 0.423 NA NA NA 0.07 0.0982 ...\n##  $ bodywt      : num  50 0.48 1.35 0.019 600 ...<\/code><\/pre>\n
Let\u2019s say we have written a groundbreaking paper on the relationship between body size and sleep time. Therefore, we want to present a plot of the log of body weight by the total sleep time<\/p>\n
sleepplot = ggplot<\/span>(data =<\/span> msleep, aes<\/span>(x =<\/span> log<\/span>(bodywt), y =<\/span> sleep_total)) + geom_point<\/span>(aes<\/span>(color =<\/span> vore))\nsleepplot<\/code><\/pre>\n
\"\"<\/figure>\n
We made a beautiful model of this relationship<\/p>\n
slp = lm<\/span>(sleep_total ~ log<\/span>(bodywt), data =<\/span> msleep)\nsummary<\/span>(slp)<\/code><\/pre>\n
## \n## Call:\n## lm(formula = sleep_total ~ log(bodywt), data = msleep)\n## \n## Residuals:\n##    Min     1Q Median     3Q    Max \n## -6.499 -2.567 -0.168  2.047 10.193 \n## \n## Coefficients:\n##             Estimate Std. Error t value Pr(>|t|)    \n## (Intercept)   11.089      0.418   26.54   <2e-16 ***\n## log(bodywt)   -0.777      0.125   -6.22    2e-08 ***\n## ---\n## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1\n## \n## Residual standard error: 3.68 on 81 degrees of freedom\n## Multiple R-squared:  0.323,  Adjusted R-squared:  0.315 \n## F-statistic: 38.7 on 1 and 81 DF,  p-value: 2.05e-08<\/code><\/pre>\n
Let\u2019s put the model on the plot<\/p>\n
sleepplot = sleepplot + geom_abline<\/span>(intercept =<\/span> coef<\/span>(slp)[1<\/span>], slope =<\/span> coef<\/span>(slp)[2<\/span>])\nsleepplot<\/code><\/pre>\n
\"\"<\/figure>\n
It\u2019s beautiful! I love it! Unfortunately, you want to submit to Science (you might as well aim high), and this is what they say about figures:http:\/\/www.sciencemag.org\/site\/feature\/contribinfo\/prep\/prep_subfigs.xhtml<\/a><\/p>\n
So we have several problems:<\/p>\n