arabidopsiscode2
UCL student
2025-11-10
Reading in the file:
library(readxl)## Warning: package 'readxl' was built under R version 4.4.3arabidopsis <- read_excel("arabidopsisdata.xlsx")## New names:
## • `` -> `...7`arabidopsis## # A tibble: 499 × 7
## SaltConc Session Variety RootLength LateralRootNumber LeafNumber ...7
## <dbl> <dbl> <chr> <dbl> <dbl> <dbl> <chr>
## 1 0 2 Cvi-0 0 0 0 <NA>
## 2 0 2 Cvi-0 0 0 0 <NA>
## 3 0 2 Cvi-0 0 0 0 <NA>
## 4 0 2 Cvi-0 0 0 0 <NA>
## 5 0 2 Cvi-0 0 0 0 <NA>
## 6 0 2 Cvi-0 0 0 0 <NA>
## 7 0 2 Cvi-0 0 0 0 <NA>
## 8 0 2 Cvi-0 0 0 0 <NA>
## 9 0 2 Cvi-0 0 0 0 <NA>
## 10 0 2 Cvi-0 0 0 0 <NA>
## # ℹ 489 more rowslibrary(ggplot2)## Warning: package 'ggplot2' was built under R version 4.4.3I was having some trouble with the variable type, so this is just telling R to consider each concentration of salt as a unique factor.
unique(arabidopsis$SaltConc)## [1] 0 80 160head(arabidopsis$SaltConc, 10)## [1] 0 0 0 0 0 0 0 0 0 0str(arabidopsis$SaltConc)## num [1:499] 0 0 0 0 0 0 0 0 0 0 ...arabidopsis$SaltConc <- factor(arabidopsis$SaltConc,
levels = c(0, 80, 160),
labels = c("0 mM", "80 mM", "160 mM"))
unique(arabidopsis$SaltConc)## [1] 0 mM 80 mM 160 mM
## Levels: 0 mM 80 mM 160 mMMaking a subset of the fourth session of the experiment containing the root lengths and lateral root numbers of all A. thaliana samples.
arabidopsissession4 <- subset(arabidopsis, Session == 4)
arabidopsissession4## # A tibble: 148 × 7
## SaltConc Session Variety RootLength LateralRootNumber LeafNumber ...7
## <fct> <dbl> <chr> <dbl> <dbl> <dbl> <chr>
## 1 0 mM 4 Cvi-0 0 0 0 <NA>
## 2 0 mM 4 Cvi-0 0 0 0 <NA>
## 3 0 mM 4 Cvi-0 0 0 0 <NA>
## 4 0 mM 4 Cvi-0 0 0 0 <NA>
## 5 0 mM 4 Cvi-0 0 0 0 <NA>
## 6 0 mM 4 Cvi-0 0 0 0 <NA>
## 7 0 mM 4 Cvi-0 0 0 0 <NA>
## 8 0 mM 4 Cvi-0 0 0 0 <NA>
## 9 0 mM 4 Cvi-0 0 0 0 <NA>
## 10 0 mM 4 Cvi-0 0 0 0 <NA>
## # ℹ 138 more rowsThis is using ggplot to create a box plot of all the root lengths combined. The ‘patchwork’ tool combines the three separately created graphs into one for convenience.
names(arabidopsissession4) <- make.names(names(arabidopsissession4))
library(ggplot2)
library(dplyr)## Warning: package 'dplyr' was built under R version 4.4.3##
## Attaching package: 'dplyr'## The following objects are masked from 'package:stats':
##
## filter, lag## The following objects are masked from 'package:base':
##
## intersect, setdiff, setequal, union#these are boxplots that separate the root lengths by salt concentration.
rootlength0 <- ggplot(filter(arabidopsissession4, SaltConc == "0 mM"),
aes(x = Variety, y = RootLength, fill = Variety)) +
geom_boxplot(fill = "#F8766D", alpha = 0.8) +
labs(title = "0 mM", x = "A. thaliana varieties", y = "Root length (mm)") +
theme_minimal(base_size = 13) +
theme(axis.text.x = element_text(angle = 45, hjust = 1),
legend.position = "none")
rootlength80 <- ggplot(filter(arabidopsissession4, SaltConc == "80 mM"),
aes(x = Variety, y = RootLength, fill = Variety)) +
geom_boxplot(fill = "#7CAE00", alpha = 0.8) +
labs(title = "80 mM", x = "A. thaliana varieties", y = "Root length (mm)") +
theme_minimal(base_size = 13) +
theme(axis.text.x = element_text(angle = 45, hjust = 1),
legend.position = "none")
rootlength160 <- ggplot(filter(arabidopsissession4, SaltConc == "160 mM"),
aes(x = Variety, y = RootLength, fill = Variety)) +
geom_boxplot(fill = "#00BFC4", alpha = 0.8) +
labs(title = "160 mM", x = "A. thaliana varieties", y = "Root length (mm)") +
theme_minimal(base_size = 13) +
theme(axis.text.x = element_text(angle = 45, hjust = 1),
legend.position = "none")
#this function adds the figures into one figure - need patchwork package
library(patchwork)## Warning: package 'patchwork' was built under R version 4.4.3(rootlength0 + rootlength80 + rootlength160) +
plot_annotation(
title = "Root Length Across Three Salt Concentrations"
) &
theme_gray(base_size = 13)
#its better to create 2 separate boxplots because it is easier to see the differences and spread of the root lengths at 160 mM, otherwise it's difficult to see them when added to the root lengths at 0 mMThese are the boxplots for lateral root growth. I chose to include in this markdown document the lateral root growth under 160 mM NaCl as proof that the data I had was less informative that the 80 mM NaCl data.
#boxplot of lateral root lengths
library(readxl)
library(ggplot2)
library(dplyr)
lr0 <- ggplot(filter(arabidopsissession4, SaltConc == "0 mM"),
aes(x = Variety, y = LateralRootNumber, fill = Variety)) +
geom_boxplot(fill = "#F8766D", alpha = 0.8) +
labs(title = "Number of lateral roots of 6 varieties at a salt concentration of 0 mM", x = "A. thaliana varieties", y = "Number of lateral roots") +
theme_minimal(base_size = 13) +
theme(axis.text.x = element_text(angle = 45, hjust = 1),
legend.position = "none")
lr80 <- ggplot(filter(arabidopsissession4, SaltConc == "80 mM"),
aes(x = Variety, y = LateralRootNumber, fill = Variety)) +
geom_boxplot(fill = "#7CAE00", alpha = 0.8) +
labs(title = "Number of lateral roots of 6 varieties at a salt concentration of 80 mM", x = "A. thaliana varieties", y = "Number of lateral roots") +
theme_minimal(base_size = 13) +
theme(axis.text.x = element_text(angle = 45, hjust = 1),
legend.position = "none")
lr160 <- ggplot(filter(arabidopsissession4, SaltConc == "160 mM"),
aes(x = Variety, y = LateralRootNumber, fill = Variety)) +
geom_boxplot(fill = "#00BFC4", alpha = 0.8) +
labs(title = "Number of lateral roots of 6 varieties at a salt concentration of 0 mM", x = "A. thaliana varieties", y = "Number of lateral roots") +
theme_minimal(base_size = 13) +
theme(axis.text.x = element_text(angle = 45, hjust = 1),
legend.position = "none")
lr160
#clearly there's no growth, so this is pointless to look at. we have added too much salt for lateral root growth to occur. use data for 80 mM and exclude this graph.
arabidopsissession4 %>%
filter(SaltConc %in% c("0 mM", "80 mM")) %>%
ggplot(aes(x = Variety, y = LateralRootNumber, fill = SaltConc)) +
geom_boxplot(alpha = 0.8) +
scale_fill_manual(values = c("0 mM" = "#F8766D",
"80 mM" = "#7CAE00")) +
facet_wrap(~ SaltConc, ncol = 2) +
labs(title = "Lateral root numbers at 0 mM and 80 mM NaCl",
x = "A. thaliana varieties",
y = "Lateral root number") +
theme_minimal(base_size = 13) +
theme(axis.text.x = element_text(angle = 45, hjust = 1),
legend.position = "none")
#this is a boxplot for just 0mm and 80mm Creating the scatter plot of germination rates across the experimental sessions and varieties. I chose to create a separate dataset for convenience.
#creating the scatter plot for the germination rate data over sessions 1-4
germination <- read_excel("germinationdata.xlsx")
class(germination$Session)## [1] "numeric"class(germination$SaltConc)## [1] "numeric"SaltConc <- as.factor(germination$SaltConc)
Variety <- as.factor(germination$Variety)
germination <- germination %>%
mutate(
SaltConc = factor(SaltConc, levels = c(0, 80, 160),
labels = c("0 mM", "80 mM", "160 mM")),
Variety = as.factor(Variety)
)
#reading in the file and converting some variables to factors for data visualization
ggplot(germination, aes(x = Session,
y = GerminationRate,
color = SaltConc,
linetype = SaltConc,
group = SaltConc)) +
geom_point(size = 2) +
geom_line(linewidth = 0.9) +
facet_wrap(~ Variety, ncol = 3) +
scale_color_manual(values = c("0 mM" = "#F8766D",
"80 mM" = "#7CAE00",
"160 mM" = "#00BFC4")) +
scale_linetype_manual(values = c("0 mM" = "solid",
"80 mM" = "dashed",
"160 mM" = "dotdash")) +
labs(title = "Germination rate under 0, 80, and 160 mM NaCl",
x = "Week", y = "Germination rate (%)",
color = "Salt concentration",
linetype = "Salt concentration") +
theme_gray(base_size = 13)
As the first part of my data analysis, I conducted KW tests between varieties for root length. I chose to separate the master dataset by salt concentration. Important to note that I had removed the Cvi-0 from the KW tests, as no Cvi-0 seeds had germinated in any of the plates. Comparing ecotypes to a variety whose properties contain only 0 would remove any truly significant effects between the varieties.
library(FSA)## Warning: package 'FSA' was built under R version 4.4.3## ## FSA v0.10.0. See citation('FSA') if used in publication.
## ## Run fishR() for related website and fishR('IFAR') for related book.library(dunn.test)
zerosalt <- read_excel("arabidopsisdata.xlsx", sheet = "0mM")
zerosalt## # A tibble: 50 × 6
## SaltConc Session Variety RootLength LateralRootNumber LeafNumber
## <dbl> <dbl> <chr> <dbl> <dbl> <dbl>
## 1 0 4 Pa-1 50 0 4
## 2 0 4 Pa-1 45 1 4
## 3 0 4 Pa-1 53 3 6
## 4 0 4 Pa-1 58 5 6
## 5 0 4 Pa-1 40 2 6
## 6 0 4 Pa-1 50 2 4
## 7 0 4 Pa-1 57 7 6
## 8 0 4 Pa-1 57 6 6
## 9 0 4 Pa-1 60 1 1
## 10 0 4 Pa-1 11 9 6
## # ℹ 40 more rowszerosaltkruskalwallisrootlength <- kruskal.test(RootLength~Variety, data = zerosalt)
zerosaltkruskalwallisrootlength##
## Kruskal-Wallis rank sum test
##
## data: RootLength by Variety
## Kruskal-Wallis chi-squared = 35.433, df = 4, p-value = 3.785e-07dunnTest(RootLength ~ Variety,
data = zerosalt,
method = "bh")## Warning: Variety was coerced to a factor.## Dunn (1964) Kruskal-Wallis multiple comparison## p-values adjusted with the Benjamini-Hochberg method.## Comparison Z P.unadj P.adj
## 1 Col-0 - Kew-1 3.2537589 1.138889e-03 2.847222e-03
## 2 Col-0 - Oy-0 -1.8612870 6.270365e-02 8.957665e-02
## 3 Kew-1 - Oy-0 -4.9984609 5.778973e-07 5.778973e-06
## 4 Col-0 - Pa-1 2.4149312 1.573818e-02 2.623030e-02
## 5 Kew-1 - Pa-1 -1.1973597 2.311664e-01 2.568515e-01
## 6 Oy-0 - Pa-1 4.4254010 9.626330e-06 4.813165e-05
## 7 Col-0 - Tu-PK7 -0.4833969 6.288139e-01 6.288139e-01
## 8 Kew-1 - Tu-PK7 -3.6924345 2.221176e-04 7.403920e-04
## 9 Oy-0 - Tu-PK7 1.3665145 1.717775e-01 2.147219e-01
## 10 Pa-1 - Tu-PK7 -2.9198228 3.502304e-03 7.004609e-03eightysalt <- read_excel("arabidopsisdata.xlsx", sheet = "80mM")
eightysalt## # A tibble: 45 × 6
## SaltConc Session Variety RootLength LateralRootNumber LeafNumber
## <dbl> <dbl> <chr> <dbl> <dbl> <dbl>
## 1 80 4 Pa-1 19 4 NA
## 2 80 4 Pa-1 0 0 NA
## 3 80 4 Pa-1 0 0 NA
## 4 80 4 Pa-1 0 0 NA
## 5 80 4 Pa-1 24 2 NA
## 6 80 4 Pa-1 0 0 NA
## 7 80 4 Pa-1 0 0 NA
## 8 80 4 Pa-1 26 7 NA
## 9 80 4 Pa-1 0 0 NA
## 10 80 4 Pa-1 0 0 NA
## # ℹ 35 more rowseightysaltkruskalwallisrootlength <- kruskal.test(RootLength~Variety, data = eightysalt)
dunnTest(RootLength ~ Variety,
data = eightysalt,
method = "bh")## Warning: Variety was coerced to a factor.## Dunn (1964) Kruskal-Wallis multiple comparison
## p-values adjusted with the Benjamini-Hochberg method.## Comparison Z P.unadj P.adj
## 1 Col-0 - Kew-1 0.9872659 3.235123e-01 3.594581e-01
## 2 Col-0 - Oy-0 -1.4221555 1.549811e-01 1.937264e-01
## 3 Kew-1 - Oy-0 -1.8247646 6.803655e-02 1.133943e-01
## 4 Col-0 - Pa-1 2.4106850 1.592259e-02 3.184519e-02
## 5 Kew-1 - Pa-1 0.4335832 6.645912e-01 6.645912e-01
## 6 Oy-0 - Pa-1 4.1427090 3.432273e-05 1.716137e-04
## 7 Col-0 - Tu-PK7 -2.6811026 7.338001e-03 2.446000e-02
## 8 Kew-1 - Tu-PK7 -2.5867177 9.689495e-03 2.422374e-02
## 9 Oy-0 - Tu-PK7 -1.5077565 1.316169e-01 1.880241e-01
## 10 Pa-1 - Tu-PK7 -5.2313145 1.683089e-07 1.683089e-06onehundredsixtysalt <- read_excel("arabidopsisdata.xlsx", sheet = "160mM")## New names:
## • `` -> `...7`onehundredsixtysalt## # A tibble: 28 × 7
## SaltConc Session Variety RootLength LateralRootNumber LeafNumber ...7
## <dbl> <dbl> <chr> <dbl> <dbl> <dbl> <dbl>
## 1 160 4 Pa-1 4 0.0001 0.0001 NA
## 2 160 4 Pa-1 3 0.0001 0.0001 NA
## 3 160 4 Tu-PK7 12 0.0001 0.0001 NA
## 4 160 4 Tu-PK7 12 0.0001 0.0001 NA
## 5 160 4 Tu-PK7 2 0.0001 0.0001 NA
## 6 160 4 Tu-PK7 6 0.0001 0.0001 NA
## 7 160 4 Tu-PK7 3 0.0001 0.0001 NA
## 8 160 4 Tu-PK7 2 0.0001 0.0001 NA
## 9 160 4 Tu-PK7 2 0.0001 0.0001 NA
## 10 160 4 Tu-PK7 2 0.0001 0.0001 3
## # ℹ 18 more rowsonehundredsixtysaltkruskalwallisrootlength <- kruskal.test(RootLength~Variety, data = onehundredsixtysalt)
dunnTest(RootLength ~ Variety,
data = onehundredsixtysalt,
method = "bh")## Warning: Variety was coerced to a factor.## Dunn (1964) Kruskal-Wallis multiple comparison
## p-values adjusted with the Benjamini-Hochberg method.## Comparison Z P.unadj P.adj
## 1 Col-0 - Kew-1 2.1462575 0.03185244 0.3185244
## 2 Col-0 - Oy-0 2.1093250 0.03491654 0.1745827
## 3 Kew-1 - Oy-0 -1.1945819 0.23225042 0.3870840
## 4 Col-0 - Pa-1 0.4271411 0.66927658 0.7436406
## 5 Kew-1 - Pa-1 -1.6080605 0.10782193 0.2695548
## 6 Oy-0 - Pa-1 -0.9389333 0.34776501 0.4968072
## 7 Col-0 - Tu-PK7 0.8309625 0.40599479 0.5074935
## 8 Kew-1 - Tu-PK7 -1.7625363 0.07797873 0.2599291
## 9 Oy-0 - Tu-PK7 -1.3380787 0.18087078 0.3617416
## 10 Pa-1 - Tu-PK7 0.1265061 0.89933134 0.8993313The same was done as above, but for lateral root counts.
#LATERAL ROOT NUMBER
zerosaltkruskalwallisrootnumber <- kruskal.test(LateralRootNumber~Variety, data = zerosalt)
dunnTest(LateralRootNumber ~ Variety,
data = zerosalt,
method = "bh")## Warning: Variety was coerced to a factor.## Dunn (1964) Kruskal-Wallis multiple comparison## p-values adjusted with the Benjamini-Hochberg method.## Comparison Z P.unadj P.adj
## 1 Col-0 - Kew-1 3.3068454 9.435294e-04 2.358823e-03
## 2 Col-0 - Oy-0 1.7297348 8.367766e-02 9.297518e-02
## 3 Kew-1 - Oy-0 -1.8073773 7.070347e-02 1.010050e-01
## 4 Col-0 - Pa-1 1.8748365 6.081522e-02 1.013587e-01
## 5 Kew-1 - Pa-1 -1.7386104 8.210331e-02 1.026291e-01
## 6 Oy-0 - Pa-1 0.1125511 9.103864e-01 9.103864e-01
## 7 Col-0 - Tu-PK7 -2.2110323 2.703360e-02 5.406720e-02
## 8 Kew-1 - Tu-PK7 -5.3133246 1.076431e-07 1.076431e-06
## 9 Oy-0 - Tu-PK7 -3.9927986 6.529804e-05 2.176601e-04
## 10 Pa-1 - Tu-PK7 -4.1841845 2.861917e-05 1.430958e-04eightysaltkruskalwallisrootnumber <- kruskal.test(LateralRootNumber~Variety, data = eightysalt)
eightysaltkruskalwallisrootnumber##
## Kruskal-Wallis rank sum test
##
## data: LateralRootNumber by Variety
## Kruskal-Wallis chi-squared = 33.88, df = 4, p-value = 7.886e-07dunnTest(LateralRootNumber ~ Variety,
data = eightysalt,
method = "bh")## Warning: Variety was coerced to a factor.## Dunn (1964) Kruskal-Wallis multiple comparison
## p-values adjusted with the Benjamini-Hochberg method.## Comparison Z P.unadj P.adj
## 1 Col-0 - Kew-1 1.7314923 8.336400e-02 1.190914e-01
## 2 Col-0 - Oy-0 -2.6344094 8.428380e-03 2.107095e-02
## 3 Kew-1 - Oy-0 -3.2795573 1.039701e-03 3.465670e-03
## 4 Col-0 - Pa-1 2.3539793 1.857364e-02 3.714729e-02
## 5 Kew-1 - Pa-1 -0.3511388 7.254843e-01 7.254843e-01
## 6 Oy-0 - Pa-1 5.3307046 9.783244e-08 9.783244e-07
## 7 Col-0 - Tu-PK7 0.6114232 5.409194e-01 6.010216e-01
## 8 Kew-1 - Tu-PK7 -1.3847725 1.661220e-01 2.076525e-01
## 9 Oy-0 - Tu-PK7 3.3969528 6.814071e-04 3.407035e-03
## 10 Pa-1 - Tu-PK7 -1.7903062 7.340469e-02 1.223412e-01onehundredsixtysaltkruskalwallisrootnumber <- kruskal.test(LateralRootNumber~Variety, data = onehundredsixtysalt)
dunnTest(LateralRootNumber ~ Variety,
data = onehundredsixtysalt,
method = "bh")## Warning: Variety was coerced to a factor.## Dunn (1964) Kruskal-Wallis multiple comparison
## p-values adjusted with the Benjamini-Hochberg method.## Comparison Z P.unadj P.adj
## 1 Col-0 - Kew-1 0.0000000 1.0000000 1.0000000
## 2 Col-0 - Oy-0 -1.3659791 0.1719455 0.8597277
## 3 Kew-1 - Oy-0 -0.6637465 0.5068526 1.0000000
## 4 Col-0 - Pa-1 0.0000000 1.0000000 1.0000000
## 5 Kew-1 - Pa-1 0.0000000 1.0000000 1.0000000
## 6 Oy-0 - Pa-1 0.9018539 0.3671345 1.0000000
## 7 Col-0 - Tu-PK7 0.0000000 1.0000000 1.0000000
## 8 Kew-1 - Tu-PK7 0.0000000 1.0000000 1.0000000
## 9 Oy-0 - Tu-PK7 1.4919734 0.1357061 1.0000000
## 10 Pa-1 - Tu-PK7 0.0000000 1.0000000 1.0000000Here I conducted Kruskal Wallis tests on root lengths within varieties. From the master dataset containing all the data for lateral root growth and root length for all sessions across all concentrations, I created 5 different sheets for every separate ecotype for convenience. These Kruskal Wallis tests were conducted between the concentrations of NaCl for each variety. I was most interested in Tu-PK-7 and Oy-0 as salt tolerant varieties, so I used KW tests to see if their root lengths or lateral root counts changed with concentration of NaCl.
tupk7 <- read_excel("arabidopsisdata.xlsx", sheet = "Tu-PK7")
tupk7kruskalwallis <- kruskal.test(RootLength~SaltConc, data = tupk7)
tupk7kruskalwallis##
## Kruskal-Wallis rank sum test
##
## data: RootLength by SaltConc
## Kruskal-Wallis chi-squared = 17.052, df = 2, p-value = 0.0001982dunnTest(RootLength ~ as.factor(SaltConc),
data = tupk7,
method = "bh")## Dunn (1964) Kruskal-Wallis multiple comparison## p-values adjusted with the Benjamini-Hochberg method.## Comparison Z P.unadj P.adj
## 1 0 - 160 3.2876399 1.010310e-03 0.0015154646
## 2 0 - 80 -0.6538254 5.132243e-01 0.5132243116
## 3 160 - 80 -3.9040724 9.458747e-05 0.0002837624tupk7lateralroot <- kruskal.test(LateralRootNumber ~ SaltConc, data = tupk7)
tupk7lateralroot##
## Kruskal-Wallis rank sum test
##
## data: LateralRootNumber by SaltConc
## Kruskal-Wallis chi-squared = 24.104, df = 2, p-value = 5.831e-06dunnTest(LateralRootNumber ~ as.factor(SaltConc),
data = tupk7,
method = "bh")## Dunn (1964) Kruskal-Wallis multiple comparison
## p-values adjusted with the Benjamini-Hochberg method.## Comparison Z P.unadj P.adj
## 1 0 - 160 4.892255 9.968729e-07 2.990619e-06
## 2 0 - 80 2.670212 7.580342e-03 1.137051e-02
## 3 160 - 80 -2.374755 1.756060e-02 1.756060e-02oy0 <- read_excel("arabidopsisdata.xlsx", sheet = "Oy-0")## New names:
## • `` -> `...7`oy0kruskalwallis <- kruskal.test(RootLength~SaltConc, data = oy0)
oy0kruskalwallis##
## Kruskal-Wallis rank sum test
##
## data: RootLength by SaltConc
## Kruskal-Wallis chi-squared = 31.097, df = 2, p-value = 1.768e-07oy0lateralroot <- kruskal.test(LateralRootNumber~SaltConc, data = oy0)
oy0lateralroot##
## Kruskal-Wallis rank sum test
##
## data: LateralRootNumber by SaltConc
## Kruskal-Wallis chi-squared = 31.271, df = 2, p-value = 1.62e-07dunnTest(LateralRootNumber ~ as.factor(SaltConc),
data = oy0,
method = "bh")## Dunn (1964) Kruskal-Wallis multiple comparison
## p-values adjusted with the Benjamini-Hochberg method.## Comparison Z P.unadj P.adj
## 1 0 - 160 2.394274 1.665329e-02 1.665329e-02
## 2 0 - 80 -3.021143 2.518226e-03 3.777338e-03
## 3 160 - 80 -5.555005 2.776046e-08 8.328138e-08