R-我该如何绘制这样的ggplot2

Question

I have data that looks like this:

 Name     Odds_Ratio_(log2)     p-value

 Ann          -1.80494           0.05
 Lucy          2.51017           0.1  
 Sally        -1.97779           0.01  
 ...

And I want to make graph that would look something like Figure B or C.
To have p-values on y-axis, odds ratio on x axis (divided by such vertical line) and Names next to plotted symbol. I don't have fractional size so it's impossible to change symbol size here.
How can I do it?
I really hope that someone will help me with this, as my R plotting knowledge lets me to make plots only like this:

plot(data$V2,data$V3, main="Ratio", xlab="p-value",ylab="Odds ratio (Log2)") 

.

Answer 1

You should refer to ?plot , ?abline , and ?text . That said, here's one way to do it with base functions.

d <- data.frame(Name=LETTERS, Odds_Ratio_log2=runif(26, -8, 8), 
                p_value=runif(26))
plot(d$Odds_Ratio_log2, d$p_value, pch=20, xlim=c(-8, 8), ylim=c(0, 1),
     axes=F, xlab='', ylab='', yaxs='i')
abline(v=0, lwd=3)
axis(3, lwd=3, at=seq(-8, 8, 1), cex.axis=0.8, lwd.ticks=1)
mtext('Odds (Log2)', 3, line=2.5)
text(d$Odds_Ratio_log2, d$p_value, d$Name, pos=4, cex=0.7)

---

UPDATE:

I've added yaxs='i' to the plot call above, because unless we do this, the top axis intercepts slightly higher than y=1, perhaps giving an impression of lower p-values.

Note that the plot is not "rotated", per se . Rather, we suppress axis 1 and 2 (the default x and y axes), with axes=F , and instead we plot axis 3 (the top axis; see ?axis ). We then use mtext to plot the label for axis 3. To plot a label at the left edge of the plot, you can either use the ylab argument (which I've set equal to '' ) in the plot function, or use mtext('p-value', 2) .

Alternatively, if you want the central vertical line to have ticks and labels corresponding to y-values, . Then, use segments to add tick marks, and text to add tick labels, eg:

segments(-0.1, seq(0, 1, 0.1), 0, seq(0, 1, 0.1), lwd=2)
text(rep(0, 10), seq(0, 1, 0.1), seq(0, 1, 0.1), cex=0.7, pos=2)

The end result will look something like this:

opar <- par(no.readonly = TRUE)
d <- data.frame(Name=LETTERS, Odds_Ratio_log2=runif(26, -8, 8), 
                p_value=runif(26))
plot(d$Odds_Ratio_log2, d$p_value, pch=20, xlim=c(-8, 8), ylim=c(0, 1),
     axes=F, xlab='', ylab='', yaxs='i', col='gray20')
abline(v=0, lwd=3)
axis(3, lwd=3, at=seq(-8, 8, 1), cex.axis=0.8, lwd.ticks=1)
mtext('Odds (Log2)', 3, line=2.5)
text(d$Odds_Ratio_log2, d$p_value, d$Name, pos=4, offset=0.3, cex=0.7)

par(xpd=NA)
segments(-0.1, seq(0.1, 0.9, 0.1), 0, seq(0.1, 0.9, 0.1), lwd=2)
text(rep(0, 10), seq(0.1, 0.9, 0.1), seq(0.1, 0.9, 0.1), 
     cex=0.7, pos=2, offset=0.3)
par(opar)