Learning Objective

• Create personal themes for figures to allow for flexible pivoting between journals

Pivoting Publications

When submitting a manuscript for publication, we often aim for the highest impact journal to which we feel there is a decent probability of acceptance. However, often our manuscripts are not accepted at the first choice journal, and we need to pivot the format and figures to meet the guidelines of the next chosen journal. We will discuss coding features which can ease the resubmission process, many of which we have introduced previously, including:

• Creating a personal theme
• Creating a personal color palette

• Incorporating variables for plot features likely to change during the resubmission process. Plotting features likely to change between journals include:

  • text font type
  • text font style
  • text size
  • figure label designation (A, Figure B, FigC, d., etc.)
  • figure resolution
  • figure size
  • output file type (tiff, svg, eps, etc.)

Altering the theme

Often different journals will have differing stylistic requirements pertaining the the thematic elements of the plot. We recommend creating a personal theme function to allow easy transitioning to a different set of requirements. Let’s explore this a bit more. We’ll start by viewing our current personal_theme().

View(personal_theme)

function(){ 
  theme_bw() +
    theme(axis.title = element_text(size = rel(1.25)),
          axis.text = element_text(size = rel(1.15)),
          plot.title = element_text(size = rel(1.5),
                                    hjust = 0.5),
          panel.grid = element_blank(),
          legend.position = "none")
}

We can add to this personal theme to ensure we are meeting the requirements for the journals. For instance, if we wanted to apply to the journal Science, we would need:

• sans-serif font with Helvectica if possible
• font size not smaller than 5 points
• line width minimum of 0.5 pt

We could add our preferred text font and size (greater than the minimum listed in the journal’s requirements) to our personal theme.

# Adding journal requirements to theme
personal_theme <- function(){ 
  theme_bw() +
    theme(text = element_text(family = "Helvetica",
                              size = 6),
          line = element_line(size = 0.5),
          axis.title = element_text(size = rel(1.25)),
          axis.text = element_text(size = rel(1.15)),
          plot.title = element_text(size = rel(1.5),
                                    hjust = 0.5),
          panel.grid = element_blank(),
          legend.position = "none")
}

Alternatively, if we knew these were variables likely to change for resubmission, we could code for them as variables, such as:

# Adding journal requirements to theme with variables
# Define likely to change variables at beginning of plotting script
font <- "Helvetica"
text_size <- 6
line_size <- 0.5

# Create personal theme with variables
personal_theme <- function(){ 
  theme_bw() +
    theme(text = element_text(family = font,
                              size = text_size),
          line = element_line(size = line_size),
          axis.title = element_text(size = rel(1.25)),
          axis.text = element_text(size = rel(1.15)),
          plot.title = element_text(size = rel(1.5),
                                    hjust = 0.5),
          panel.grid = element_blank(),
          legend.position = "none")
}

Altering personal color palettes

Usually, a color-blind friendly palette comprised of differing hues should be acceptable to most journals. However, if we had followed the manuscript and not changed to a color-blind friendly palette, there are many journals that would not have accepted our color choices. For instance, the journal Science among others has the following requirements:

• Avoid using red and green together
• Do not use colors that are close in hue

Luckily we have achieved both of these requirements by following best practices for color choices; however this is something to be aware of. We could easily pivot to a different color palette if we had created a personal palette containing our colors for each type of plot. Even though we used a ggplot2 layer for the viridis colors, it may be easier to pivot if you defined your personal palette to use. For instance, if we wanted to change our color palette we could easily change the palette:

# Define new palette
personal_boxplot_palette <- c("#440154FF", "#228C8DFF", "#FDE725FF")

# ggplot2 code to remain the same
ggplot(pax6_exp) +
  geom_boxplot(aes(x=group, 
                   y=normalized_counts, 
                   fill=group)) +
  ggtitle("Pax6") +
  personal_theme() +
  theme(axis.text.x = element_text(angle = 45, 
                                   vjust = 1, 
                                   hjust = 1)) +
  scale_x_discrete(name = "",
                   labels=c("Pax6:WT" = "Radial glia",
                            "neg:WT" = "Neurons", 
                            "Tbr2:WT" = "Progenitors")) +
  scale_y_continuous(name = "Normalized counts") +
  scale_fill_manual(values = personal_boxplot_palette)

Incorporating variables for plot features likely to change

We can address some of the features likely to change within the personal theme, such as font type, style and size. However, not included within the theme elements are features such as figure labels, resolution, size and output file type. For these elements, it can be helpful to create variables to specify their values so that we can easily change.

For instance, Science requires a low resolution for initial submission, then a high resolution for the publication-ready manuscript. You could easily pivot your figure if you have the resolution stored as a variable which is used for all of your plots. We show below how we can define variables for selected plot features likely to change between journals, then explore how these variables can be incorporated into the code.

# Define likely to change features as variables at beginning of the plotting script
# Theme elements
font <- "Helvetica"
text_size <- 6
line_size <- 0.5

# Figure elements
res <- 300
full_width <- 8.5
full_height <- 13
meas_units <- "in"
output_filetype <- "svg"
fig_lab_size <- 9
fig_lab_pos_x <- 0
fig_lab_pos_y <- 1
label_type <- "AUTO"

We can then incorporate these variables throughout the manuscript.

# Example code using pre-defined variables

# Create annotations for all desired volcano plots, and save them to variables
volcano_panel1 <- ggdraw(volcano_RG) + 
  draw_label("664 genes", 
             x = 0.15, 
             y = 0.82,
             hjust = 0,
             vjust = 0,
             fontface = "bold",
             size = text_size,
             fontfamily = font)

# other panels...

# Align volcano plots and label with A, B, C
volcano_grid <- plot_grid(volcano_panel1,
                          volcano_panel2,
                          volcano_panel3,
                          ncol = 3,
                          labels = label_type,
                          label_size = fig_lab_size,
                          label_x = fig_lab_pos_x,
                          label_y = fig_lab_pos_y)

# Save image
ggsave(plot = volcano_grid,
       filename = "results/volcano_figure.png",
       width = full_width,
       height = full_height/2.5,
       units = meas_units,
       dpi = res)
             

Using these three techniques can make pivoting publications much less labor intensive. Although we can’t easily code for all possible changes reviewers might desire, these basic alterations can at least ease the pain.