Displaying LM based on p-value

Question

I'm displaying linear regression models in plots using the ggpmisc package. I only want the regression line, p-value and r2-value to be showed int the plot if the p-value is less than 0.2.

Does anyone know if there is a way to choose to display these things based on the p-value?

Here's the code for the plot:

library(ggpmisc)
library(ggplot2)

formula <- y~x

ggplot(df, aes(carbon, 
               acetone, 
               fill=soil_type)) +
  geom_smooth(method = "lm",
              formula = formula, 
              color="black") +
  geom_point(aes(shape=soil_type, 
                 color=soil_type, 
                 size=soil_type)) +
  scale_fill_manual(values=c("green3", "brown")) + 
  scale_color_manual(values=c("black", "black")) + 
  scale_shape_manual(values=c(21, 24))+
  scale_size_manual(values=c(2.7, 2.0))+
  labs(shape="soil_type", 
       color="soil_type") +
  theme_bw() +
  facet_wrap(~days, 
             ncol = 2)+
  stat_poly_eq(
    aes(label = paste(stat(adj.rr.label),
                      stat(p.value.label), 
                      sep = "*\", \"*")),
    formula = formula, 
    rr.digits = 2, 
    p.digits = 1, 
    parse = TRUE,size=3.5)

Here's the dataset:

df <- structure(list(carbon = c(1.4, 0.8, 1.6, 0.1, 0.4, 0.4, 0.4, 
1.3, 0.4, 1.1, 0.2, 1, 0.4, 0.4, 0.5, 0.8, 0.1, 0.5, 0.4, 0.6, 
1.1, 0.6, 0.2, 0.2, 0.4, 0.1, 0.3, 0.5, 1.4, 0.3, 0.3, 1.1, 0.3, 
0.7, 0.4, 0.4, 1.1, 0.1, 0.6, 1.3, 0.1, 1.6, 0.4, 0.5, 0.5, 1.2, 
0.5, 0.5, 1.4, 0.8, 1.6, 0.1, 0.4, 0.4, 0.4, 1.3, 0.4, 1.1, 0.2, 
1, 0.4, 0.4, 0.5, 0.8, 0.1, 0.5, 0.4, 0.6, 1.1, 0.6, 0.2, 0.2, 
0.4, 0.1, 0.3, 0.5, 1.4, 0.3, 0.3, 1.1, 0.3, 0.7, 0.4, 0.4, 1.1, 
0.1, 0.6, 1.3, 0.1, 1.6, 0.4, 0.5, 0.5, 1.2, 0.5, 0.5), days = c(0L, 
0L, 0L, 0L, 0L, 0L, 0L, 0L, 0L, 0L, 0L, 0L, 0L, 0L, 0L, 0L, 0L, 
0L, 0L, 0L, 0L, 0L, 0L, 0L, 0L, 0L, 0L, 0L, 0L, 0L, 0L, 0L, 0L, 
0L, 0L, 0L, 0L, 0L, 0L, 0L, 0L, 0L, 0L, 0L, 0L, 0L, 0L, 0L, 10L, 
10L, 10L, 10L, 10L, 10L, 10L, 10L, 10L, 10L, 10L, 10L, 10L, 10L, 
10L, 10L, 10L, 10L, 10L, 10L, 10L, 10L, 10L, 10L, 10L, 10L, 10L, 
10L, 10L, 10L, 10L, 10L, 10L, 10L, 10L, 10L, 10L, 10L, 10L, 10L, 
10L, 10L, 10L, 10L, 10L, 10L, 10L, 10L), soil_type = c("organic", 
"mineral", "organic", "mineral", "mineral", "mineral", "mineral", 
"organic", "mineral", "organic", "mineral", "mineral", "mineral", 
"mineral", "mineral", "mineral", "mineral", "mineral", "mineral", 
"mineral", "organic", "mineral", "mineral", "mineral", "mineral", 
"mineral", "mineral", "mineral", "organic", "mineral", "mineral", 
"organic", "mineral", "mineral", "mineral", "mineral", "organic", 
"mineral", "mineral", "organic", "mineral", "organic", "mineral", 
"mineral", "mineral", "organic", "mineral", "mineral", "organic", 
"mineral", "organic", "mineral", "mineral", "mineral", "mineral", 
"organic", "mineral", "organic", "mineral", "mineral", "mineral", 
"mineral", "mineral", "mineral", "mineral", "mineral", "mineral", 
"mineral", "organic", "mineral", "mineral", "mineral", "mineral", 
"mineral", "mineral", "mineral", "organic", "mineral", "mineral", 
"organic", "mineral", "mineral", "mineral", "mineral", "organic", 
"mineral", "mineral", "organic", "mineral", "organic", "mineral", 
"mineral", "mineral", "organic", "mineral", "mineral"), acetone = c(0.9, 
0.7, 0.5, 44.4, 44.2, 9.7, 66, 3.3, 8.6, 26.8, 111.4, 14.5, 40.7, 
23.2, 51.6, 3.5, 64.3, 9.8, 48.5, 39.4, 0.2, 24.2, 55.3, 30.2, 
28.9, 63.6, 80.7, 50.4, 0.9, 34.4, 102.2, 2.8, 16.5, 9.7, 32.1, 
124.4, 3.7, 56.8, 10.6, 0.7, 41.1, 1.3, 62.5, 1.1, 86.3, 0.1, 
2.7, 5, 0.1, 0.1, 0.1, 179.1, 60.9, 2.6, 65.3, 14.7, 0, 34.9, 
133.7, 0, 56, 36.2, 2, 0.2, 44.9, 24.5, 123.8, 26.5, 0.1, 0.2, 
23.6, 146.3, 0.3, 169.2, 164.4, 30, 0, 0, 123.1, 0.2, 3.1, 58, 
0, 0.1, 0, 44.9, 1, 0, 102.6, 3.9, 91.4, 1.3, 21.8, 0.1, 0.6, 
1.8)), row.names = c(NA, -96L), class = "data.frame")

New df

new_df <- structure(list(log10_carbon_content_pct = c(1.37049458496569, 
0.832668550451795, 1.59213788019068, 0.145507171409663, 0.446381812222442, 
0.439569517147175, 0.422589839851482, 1.28870743057217, 0.400192488592576, 
1.09659720835789, 0.241297387109993, 0.961610908091281, 0.398026858883686, 
0.392257161341674, 0.453700473359772, 0.806451323247262, 0.11544408343624, 
0.495474955889315, 0.354492600589436, 0.61526599889915, 1.14182589451108, 
0.600537294364469, 0.160768561861128, 0.180699201296035, 0.447002898466162, 
0.104657791008796, 0.276806345628763, 0.530903734802764, 1.41408715182753, 
0.272305844402086, 0.250175948083925, 1.12073840554294, 0.323045735481701, 
0.652971172017589, 0.373463721632369, 0.378942698613437, 1.13800253645643, 
0.0874264570362855, 0.601027315144485, 1.34486364979713, 0.139721704815204, 
1.60809259256346, 0.379305517750582, 0.51215053692203, 0.466496903744401, 
1.23437806425139, 0.541766399511599, 0.471365065418019, 1.37049458496569, 
0.832668550451795, 1.59213788019068, 0.145507171409663, 0.446381812222442, 
0.439569517147175, 0.400192488592576, 1.09659720835789, 0.241297387109993, 
0.961610908091281, 0.398026858883686, 0.392257161341674, 0.453700473359772, 
0.806451323247262, 0.11544408343624, 0.495474955889315, 0.354492600589436, 
0.61526599889915, 1.14182589451108, 0.600537294364469, 0.160768561861128, 
0.180699201296035, 0.447002898466162, 0.104657791008796, 0.276806345628763, 
0.530903734802764, 1.41408715182753, 0.272305844402086, 0.250175948083925, 
1.12073840554294, 0.323045735481701, 0.652971172017589, 0.373463721632369, 
0.378942698613437, 1.13800253645643, 0.0874264570362855, 0.601027315144485, 
1.34486364979713, 0.139721704815204, 1.60809259256346, 0.379305517750582, 
1.23437806425139, 0.541766399511599, 0.471365065418019), daysincubated4 = c(0L, 
0L, 0L, 0L, 0L, 0L, 0L, 0L, 0L, 0L, 0L, 0L, 0L, 0L, 0L, 0L, 0L, 
0L, 0L, 0L, 0L, 0L, 0L, 0L, 0L, 0L, 0L, 0L, 0L, 0L, 0L, 0L, 0L, 
0L, 0L, 0L, 0L, 0L, 0L, 0L, 0L, 0L, 0L, 0L, 0L, 0L, 0L, 0L, 94L, 
94L, 94L, 94L, 94L, 94L, 94L, 94L, 94L, 94L, 94L, 94L, 94L, 94L, 
94L, 94L, 94L, 94L, 94L, 94L, 94L, 94L, 94L, 94L, 94L, 94L, 94L, 
94L, 94L, 94L, 94L, 94L, 94L, 94L, 94L, 94L, 94L, 94L, 94L, 94L, 
94L, 94L, 94L, 94L), soil_type = c("organic", "mineral", "organic", 
"mineral", "mineral", "mineral", "mineral", "organic", "mineral", 
"organic", "mineral", "mineral", "mineral", "mineral", "mineral", 
"mineral", "mineral", "mineral", "mineral", "mineral", "organic", 
"mineral", "mineral", "mineral", "mineral", "mineral", "mineral", 
"mineral", "organic", "mineral", "mineral", "organic", "mineral", 
"mineral", "mineral", "mineral", "organic", "mineral", "mineral", 
"organic", "mineral", "organic", "mineral", "mineral", "mineral", 
"organic", "mineral", "mineral", "organic", "mineral", "organic", 
"mineral", "mineral", "mineral", "mineral", "organic", "mineral", 
"mineral", "mineral", "mineral", "mineral", "mineral", "mineral", 
"mineral", "mineral", "mineral", "organic", "mineral", "mineral", 
"mineral", "mineral", "mineral", "mineral", "mineral", "organic", 
"mineral", "mineral", "organic", "mineral", "mineral", "mineral", 
"mineral", "organic", "mineral", "mineral", "organic", "mineral", 
"organic", "mineral", "organic", "mineral", "mineral"), log10_acetone_c = c(0.00926846768640111, 
0.00722297480690438, 0.00476160511452692, 0.444394072789671, 
0.442046700697262, 0.0969444813777115, 0.659755819077841, 0.0331353719785704, 
0.0860951658767628, 0.268398621135863, 1.11425178483619, 0.144553061208862, 
0.406832035925707, 0.23171358353469, 0.516127469481814, 0.0348431782930108, 
0.643065970146376, 0.0976949557752846, 0.485466415318889, 0.394413062169997, 
0.00215649305508422, 0.241895264091051, 0.553024436602299, 0.302071278712034, 
0.289062005904557, 0.63551683217124, 0.806576564543876, 0.504060450046605, 
0.00875624846772753, 0.343705138058693, 1.02196979128145, 0.0275595437157376, 
0.165301609757072, 0.0968005281885054, 0.321248914837189, 1.2444284170588, 
0.0365188178564554, 0.567626974656115, 0.10556180687771, 0.00690130440471077, 
0.411060711247439, 0.0126107339499284, 0.625415265123349, 0.0107804497649973, 
0.863015429856585, 0.000900918219072745, 0.0266090695624902, 
0.0503551827004673, 0.000268027005920481, 0.000469317124751776, 
3.95824821597101e-05, 0.00441758233902833, 0.00236289505353141, 
0.00113488982479906, 0, 0, 0, 0, 0.0531046200284991, 0, 0, 0.000214196723493331, 
0.00302534713027796, 0.000436083384348923, 0.000226218782648512, 
0.00292979624099701, 0.00124508843867096, 0, 0.0286531848530279, 
0.0134556110640359, 0, 0, 0, 0.0104783275343697, 0.0154919389302452, 
0.0197930173217508, 0, 0, 0, 0, 0, 0, 0.00051087017306838, 0.000186632771190318, 
0.00500797053508424, 0, 0.318561145793628, 0, 0, 0.0126117092437348, 
0.00699751577956711, 0)), row.names = c(NA, -92L), class = "data.frame")


names = c(0,0) #Create a starting point of a matrix for the group names

#For each group, run a lm to find if pvalue < 0.2
for(i in unique(new_df$daysincubated4)){
  for(j in unique(new_df$soil_type)){
    lm = summary(lm(log10_acetone_c~log10_carbon_content_pct, new_df[new_df$daysincubated4==i & new_df$soil_type==j,])) 
    p = pf(lm$fstatistic[1], lm$fstatistic[2], lm$fstatistic[3], lower.tail=FALSE)
    if(p < 0.2){names = rbind(names, c(i,j))} #Get the groups that pass
  }
}

names = names[-1,] #Remove starting point

new_df2 = new_df[new_df$daysincubated4%in%names[,1] & new_df$soil_type%in%names[,2],]

formula <- y~x

(acetone_c_vs_cc <- ggplot(new_df, 
                           aes(log10_carbon_content_pct, 
                               log10_acetone_c, 
                               fill=soil_type)) +
    geom_smooth(method = "lm",
                formula = formula, color="black", data = new_df2) +
    geom_point(aes(shape=soil_type, color=soil_type, size=soil_type)) +
    scale_fill_manual(values=c("#00AFBB", "brown")) + 
    scale_color_manual(values=c("black", "black")) + 
    scale_shape_manual(values=c(21, 24))+
    scale_size_manual(values=c(2.4, 1.7))+
    labs(shape="soil_type", color="soil_type") +
    labs(x = "Soil organic carbon (%)", 
         y = "Emission (umol/g dw SOC/h)", 
         title = "Acetone vs Carbon content", 
         subtitle = "Emission and carbon data has been log10 transformed") + 
    theme_bw() +
    facet_wrap(~daysincubated4, 
               ncol = 4)+
    stat_poly_eq(data = new_df2,
      aes(label = paste(stat(adj.rr.label),
                        stat(p.value.label), 
                        sep = "*\", \"*")),
      formula = formula, 
      rr.digits = 2, 
      p.digits = 1, 
      parse = TRUE,size=3.5))

user_13048728 · Answer

I'm not sure if this is the best way, but you can run the lm's by hand to check your condition, then subset your df to use only the observations of the group that passes:

First, to check the condition:

names = c(0,0) #Create a starting point of a matrix for the group names

#For each group, run a lm to find if pvalue < 0.2
for(i in unique(df$days)){
  for(j in unique(df$soil_type)){
    lm = summary(lm(acetone~carbon, df[df$days==i & df$soil_type==j,])) 
    p = pf(lm$fstatistic[1], lm$fstatistic[2], lm$fstatistic[3], lower.tail=FALSE)
    if(p < 0.2){names = rbind(names, c(i,j))} #Get the groups that pass
  }
}

names = names[-1,] #Remove starting point

Now, create a subset of the df, and pass it as a data argument to both geom_smooth and stat_poly_eq:

df2 = df[df$days%in%names[,1] & df$soil_type%in%names[,2],]

ggplot(df, aes(carbon, 
                   acetone, 
                   fill=soil_type)) +
  geom_smooth(method = "lm",
              formula = formula, 
              color = "black",
              data = df2) +
  geom_point(aes(shape=soil_type, 
                 color=soil_type, 
                 size=soil_type)) +
  scale_fill_manual(values=c("green3", "brown")) + 
  scale_color_manual(values=c("black", "black")) + 
  scale_shape_manual(values=c(21, 24))+
  scale_size_manual(values=c(2.7, 2.0))+
  labs(shape="soil_type", 
       color="soil_type") +
  theme_bw() +
  facet_wrap(~days, 
             ncol = 2)+
  stat_poly_eq(
    data = df2,
    aes(label = paste(stat(adj.rr.label),
                      stat(p.value.label), 
                      sep = "*\", \"*")),
    formula = formula, 
    rr.digits = 2, 
    p.digits = 1, 
    parse = TRUE,size=3.5)

Output:

EDIT 1

The part where i subset the df was wrong (it only worked because there was a group that didn't passed at all). The problem is that df$days%in%names[,1] & df$soil_type%in%names[,2] doesn't check pair by pair. So we actually need do do a loop:

#Create subset of df with groups that passed
new_df2 = numeric()
for(i in 1:nrow(names)){
  new_df2 = rbind(new_df2,
                  new_df[new_df$daysincubated4%in%names[i,1] & new_df$soil_type%in%names[i,2],])
}

Obs: this rearranges the data frame, but i don't think this would be a problem for you.

Output:

Obs: the 0.2 p-values are actually rounded up, so they should pass.

Displaying LM based on p-value

1 Answers