Equation of a geom_smooth line

[Schisto]

I think I have calculated the straight line of a geom_smooth line in a generated figure but would like to make it tidier. See example below:

Example data called data2:

Well_numbers Sample Dilution mIU.mL Type

1	Dilution 1:200	0.005	1.4450	Control
2	Dilution 1:200	0.005	1.2905	Control
3	Dilution 1:200	0.005	1.4425	Control
4	Dilution 1:500	0.002	1.1065	Control
5	Dilution 1:500	0.002	0.9475	Control
6	Dilution 1:500	0.002	0.9955	Control

library(ggplot2)
library(Hmisc)

lm_eqn = function(data2){
    m = lm(mIU.mL ~ log10(Dilution), data2);
    eq <- substitute(italic(y) == a + b %.% italic(x)*","~~italic(r)^2~"="~r2, 
         list(a = format(coef(m)[1], digits = 4), 
              b = format(coef(m)[2], digits = 4), 
             r2 = format(summary(m)$r.squared, digits = 3)))
    as.character(as.expression(eq));                 
}

ggplot(data2, aes(log10(Dilution), mIU.mL)) +
  stat_summary(fun.y = mean, geom = "point") +
  labs(title = "My Informative Title", x = "X axis label", y = "Y axis label") +
  geom_smooth(method = "lm", se = TRUE, colour = 'green')+theme_light() + geom_text(x = -3, y = 1.2,label = lm_eqn(data2), parse = TRUE)

My output is:

The equation of the line is y = c(3.53) + c(0.9301)x

I believe this is the same as y = 0.9301x + 3.53 (or y = 3.53 + 0.9301x). This makes sense when inputting the known data.

If this is true does anyone know how I remove the c values from the equation in the graph above? Or another way to calculate the equation of the straight line generated in the graph?

Thanks in advance.

[Schisto]

Well_numbers	Sample	Dilution	mIU/mL	Type
1	Dilution 1:200	0.005	1.445	Control
2	Dilution 1:200	0.005	1.2905	Control
3	Dilution 1:200	0.005	1.4425	Control
4	Dilution 1:500	0.002	1.1065	Control
5	Dilution 1:500	0.002	0.9475	Control
6	Dilution 1:500	0.002	0.9955	Control
7	Dilution 1:2000	0.0005	0.4295	Control
8	Dilution 1:2000	0.0005	0.4635	Control
9	Dilution 1:2000	0.0005	0.4925	Control

All data for above graph just realised I'd only given the head(data 2) data previously.

[smichal]

Hi @Schisto,

if you only want to format your equation differently you could start with a string. Maybe that helps a little...

I'm using different data but it should transfer easily to your problem, I hope.

library(tidyverse)

# a nonsense fit but never mind
fit <- lm(data = mtcars, formula = mpg ~ hp + cyl)
coeffs <- fit$coefficients


# This is the equation string without italic etc. formatting
# start with the coefficients excluding "(Intercept)"
str_c(c(str_c(sprintf("%+ .2f", coeffs[-1]), names(coeffs)[-1], sep = "*"), 
        # add the intercept without a name
        sprintf("%+ .2f", coeffs[1])),
      collapse = " ")
#> [1] "-0.02*hp -2.26*cyl +36.91"

^{Created on 2020-09-21 by the reprex package (v0.3.0)}

You could also check the package latex2exp to use a TeX expression.