Package 'grafify'

Description

These data are from in vitro measurements of death of host cells (measured as percentage of total cells) after infection with three different strains of a pathogenic bacterium, from five independent experiments. The three strains are three levels within the fixed factor Genotype. The five independent experiments are levels within the random variable Experiment. These data can be analysed using linear mixed effects modelling. These data are from Goddard et al, Cell Rep, 2019, doi.org/10.1016/j.celrep.2019.03.100

Usage

data_1w_death

Format

data.frame: 15 obs. of 3 variables.

Experiment

Experiment - a random factor with 5 levels "Exp_1","Exp_2"...

Genotype

Genotypes - a fixed factor with 3 levels: "WT","KO_1","KO_2".

Death

Numerical dependent variable indicating percentage cell death.

Description

Data from Festing, ILAR Journal (2014) 55, 472–476 <doi: 10.1093/ilar/ilu045>. These data are suitable for two-way linear mixed effects modelling. The activity of GST (numerical dependent variable) was measured in 4 strains of mice (levels with the fixed factor Strain) either treated or controls (levels within the fixed factor Treatment). Once mouse each was used in two randomised blocks, which is the random factor (Block).

Usage

data_2w_Festing

Format

data.frame: 16 obs. of 4 variables:

Block

A random factor with 2 levels "A" and "B".

Treatment

A fixed factor with 2 levels: "Control" & "Treated"

Strain

A fixed factor with 4 levels: "129Ola", "A/J", "NIH" & "BALB/C"

GST

Numerical dependent variable indicating GST activity measurement

Description

These are measurements of death of infected host cells (as percentage of total cells) upon infection with two strains of bacteria, measured at two time points, in 6 independent experiments. These data repeated-measures data suitable for two-way linear mixed effects modelling with experiment and subjects as random factors.

Usage

data_2w_Tdeath

Format

data.frame: 24 obs. of 6 variables:

Experiment

A random factor with 6 levels "e1", "e2"...

Time

A fixed factor with 2 levels: "t100" & "t300".

Time2

A numeric column that allows plotting data on a quantitative "Time" axis. The "Time" column has "factor" type values that should be used for the ANOVA..

Genotype

A fixed factor with 2 levels that we want to compare "WT" & "KO".

Subject

A random factor with 12 levels: "s1", "s2"... These are cell culture wells that were measured at two time points, and indicate "subjects" that underwent repeated-measures within each of 6 experiments. Subject IDs for WT and KO are unique and clearly indicate different wells.

PI

Numerical dependent variable indicating propidium iodide dye uptake as a measure of cell death. These are percentage of dead cells out of total cells plated.

Description

An example dataset on measurements of blood cholesterol levels measured in 5 subjects measured before and after receiving a Drug. Five patients each were recruited at 5 hospitals (a-e), so that there are 25 different subjects (1-25) measured twice. Data are from Micro/Immuno Stats

Usage

data_cholesterol

Format

tibble: 30 obs. of 3 variables:

Hospital

Factor with 5 levels (a-e), representing different hospitals where subjects were recruited.

Subject

A factor with 25 levels denoting individuals on whom measurements were made twice.

Treatment

A factor with 2 levels indicating when measurements were made, i.e. before and after drug.

Cholesterol

Numerical dependent variable indicating measured doubling time in min.

Description

An example dataset showing measurements of E. coli doubling times (in min) measured by 10 different students in 3 independent experiments each. Note that Experiments are just called Exp1-Exp3 even though Exp1 of any of the students are not connected in anyway - this will confuse R! Data are from Micro/Immuno Stats

Usage

data_doubling_time

Format

tibble: 30 obs. of 3 variables:

Student

Factor with 10 levels, representing different students.

Experiment

A factor with 3 levels representing independent experiments.

Doubling_time

Numerical dependent variable indicating measured doubling time in min.

Description

An example dataset for paired difference Student's t test. These are bodyweight (Mass) in grams of same mice left untreated or treated, which are two groups to be compared. The data are in a longtable format, and the two groups are levels within the factor "Condition". The Subject column lists ID of matched mice that were measured without and with treatment. These data are from Sanchez-Garrido et al, Sci Signal, 2018, DOI: 10.1126/scisignal.aat6903.

Usage

data_t_pdiff

Format

data.frame: 20 obs. of 3 variables:

Subject

Factor with 10 levels, denoted by capital letters, representing individuals or subjects.

Condition

A fixed factor with 2 levels: "Untreated" & "Treated".

Mass

Numerical dependent variable indicating body mass of mice

Description

An example dataset for paired ratio Student's t test. These are Cytokine measurements by ELISA (in ng/ml) from 33 independent in vitro experiments performed on two Genotypes that we want to compare. The data are in a longtable format, and the two groups are levels within the factor "Genotype". The Experiment column lists ID of matched experiments.

Usage

data_t_pratio

Format

data.frame: 66 obs. of 3 variables:

Genotype

Factor with 2 levels, representing genotypes to be compared ("WT" & "KO").

Experiment

A random factor with 33 levels representing independent experiments, denoted as "Exp_1", "Exp_2"...

Cytokine

Numerical dependent variable indicating cytokine measured by ELISA.

Description

A subset of data from (Lathro RC, 2000) (doi:10.6073/pasta/ec3d0186753985147d4f283252388e05) provided by the Wisconsin Department of Natural Resources

Usage

data_zooplankton

Format

tibble: 1127 obs. of 8 variables:

day

Numeric integer variable.

year

Numeric integer variable of years during which data were collected.

lake

This data is for lake Menona; data for other others not included in this subset.

taxon

Names of zooplankton taxa as factor of 8 levels.

density

Numeric values of density of measurements.

density_adj

Numeric values of adjusted density .

min_density

Numeric values of minimum densities.

desnsity_scaled

Numeric value of scaled density.

Description

One of two functions for fitting generalised additive models (gam) with the mgcv package. It will use the gam() function in mgcv for ANOVA designs with up to two categorical fixed factors (with two or more levels; Fixed_Factor), and exactly one factor is a continuous variable (e.g. time), which is called Smooth_Factor.

1. ga_model

2. ga_anova

Usage

ga_anova(
  data,
  Y_value,
  Fixed_Factor,
  Smooth_Factor,
  Random_Factor = NULL,
  Nodes = NULL,
  ...
)

Arguments

Details

A smooth function is fitted with factor-wise smooth basis function (⁠by = ⁠). A default value for number of nodes (the argument k in gam) may work, but a specific number can be provided using the Nodes argument. The model is fit using the REML method. When two categorical fixed factors are provided, an interaction term is included for main effects and smooth basis functions.

If a Random_Factor is also provided, it is fitted using bs = "re" smooth.

Value

ANOVA table of class "anova" and "data.frame".

Examples

#with zooplankton data
ga_anova(data = data_zooplankton,
Y_value = "log(density_adj)",
Fixed_Factor = "taxon",
Smooth_Factor = "day")

Description

One of two functions for fitting generalised additive models (gam) with the mgcv package. It will use the gam() function in mgcv for ANOVA designs with up to two categorical fixed factors (with two or more levels; Fixed_Factor), and exactly one factor is a continuous variable (e.g. time), which is called Smooth_Factor.

1. ga_model

2. ga_anova

A smooth function is fitted with factor-wise smooth basis function (⁠by = ⁠). A default value for number of nodes (the argument k in gam) may work, but a specific number can be provided using the Nodes argument. The model is fit using the REML method. When two categorical fixed factors are provided, an interaction term is included for main effects and smooth basis functions.

Usage

ga_model(
  data,
  Y_value,
  Fixed_Factor,
  Smooth_Factor,
  Random_Factor = NULL,
  Nodes = "NULL",
  ...
)

Arguments

Details

If a Random_Factor is also provided, it is fitted using bs = "re" smooth.

Value

This function gives a generalised additive model object of class "gam", "lm" and "glm".

Examples

#fit a model with zooplankton data
z1 <- ga_model(data = data_zooplankton,
Y_value = "log(density_adj)",
Fixed_Factor = "taxon",
Smooth_Factor = "day")

Description

Function to make grafify colour scheme. Thank you Dr Simon.

Usage

get_graf_colours(...)

Arguments

Details

To visualise grafify colours use plot_grafify_palette.

Value

This function returns names and hexcodes of colours in grafify as a character vector.

Description

graf_col_palette and graf_col_palette_default functions generate colours for grafify scale functions. graf_col_palette picks sequential colours when the number of discrete colours needed is less than that in the palette. This is the default for grafify with ColoSeq = TRUE. If the number of colours required is more than that in the discrete palette, it fills intervening colours using the colorRampPalette[grDevices] function.

Usage

graf_col_palette(palette = "okabe_ito", reverse = FALSE, ...)

Arguments

Details

graf_col_palette_default picks the most distant colours within the palette, rather than in the sequence they are in the palette, when the number of colours required is less than that in the palette.

Colour order can be reversed in both functions.

When only one colour discreet is required, and you want to reverse the colour palette, ColSeq should be set to FALSE.

Value

This generates required number of sequential colours from the chosen grafify palette when called by scale functions of ggplot2.

Description

graf_col_palette and graf_col_palette_default functions generate colours for grafify scale functions. graf_col_palette picks sequential colours when the number of discrete colours needed is less than that in the palette. This is the default for grafify with ColoSeq = TRUE. If the number of colours required is more than that in the discrete palette, it fills intervening colours using the colorRampPalette[grDevices] function.

Usage

graf_col_palette_default(palette = "okabe_ito", reverse = FALSE, ...)

Arguments

Details

graf_col_palette_default picks the most distant colours within the palette, rather than in the sequence they are in the palette, when the number of colours required is less than that in the palette.

Colour order can be reversed in both functions.

When only one colour discreet is required, and you want to reverse the colour palette, ColSeq should be set to FALSE.

Value

This generates required number of distant colours from the chosen grafify palette when called by scale functions of ggplot2.

Description

To visualise these colours use plot_grafify_palette. okabe_ito, bright, contrast, dark, light, muted, pale, vibrant, yello_conti from Paul Tol's post. Zesty, Pastel, Elegant from this link. Colour hexcodes for fishy, kelly, r4, safe, OrBl_div, PrGn_div, blue_conti, grey_conti taken from cols4all:c4a_gui package. All schemes are colour blind-friendly.

Usage

graf_colours

Format

An object of class character of length 154.

Value

This is a character vector with names and hexcodes of colours used by palette functions. It is used by get_graf_colours to generate palettes.

Description

To visualise these colours use plot_grafify_palette.

Usage

graf_palettes

Format

An object of class list of length 18.

Value

This function returns a list of palettes in grafify with names and hexcodes of colours in those palettes. Names of palettes available are as follows:

Categorical/discreet palettes:

• okabe_ito

• bright

• contrast

• dark

• kelly

• light

• muted

• pale

• r4

• safe

• vibrant

Sequential quantitative palettes:

• grey_conti

• blue_conti

• yellow_conti

Divergent quantitative palettes:

• OrBl_div

• PrGn_div

Description

The make_1way_data, make_1way_rb_data, make_2way_data and make_2way_rb_data functions generate independent or randomised block (rb) design data of one-way or two-way designs.

Usage

make_1way_data(Group_means, Num_obs, Residual_SD)

Arguments

Details

Random variates from the normal distribution based on user provided mean and SD provided are generated. For independent designs, the Residual_SD argument is used to set expected residual SD from the linear model. Exp_SD is used to set experiment-to-experiment SD, that will be assigned to the random factor for rb designs.

Num_exp sets the number of independent measurements per group.

For one-way designs, the user provides Group_means as a vector. Number of levels are recognised based on number of means. For two-way designs, two vectors are to be provided by the user containing means of levels of a second factor. Number of means in both vectors should be the same. These functions can only handle balanced designs, i.e. same number of observations in all groups.

The output is a data frame with one or two columns denoting the fixed factor with levels that match the number of means entered. For rb data, the column for RandFac denotes levels of the blocking factor. The quantitative response variables are in the numeric Values column.

Value

This function produces a data.frame object containing simulated data.

Examples

#Basic usage with three levels within Factor_X, 
#20 observations in each group, with residual SD 15

one_independent_tab <- make_1way_data(c(350, 250, 100), 15, 20)

str(one_independent_tab)
head(one_independent_tab)

Description

The make_1way_data, make_1way_rb_data, make_2way_data and make_2way_rb_data functions generate independent or randomised block (rb) design data of one-way or two-way designs.

Usage

make_1way_rb_data(Group_means, Num_exp, Exp_SD, Residual_SD)

Arguments

Details

Random variates from the normal distribution based on user provided mean and SD provided are generated. For independent designs, the Residual_SD argument is used to set expected residual SD from the linear model. Exp_SD is used to set experiment-to-experiment SD, that will be assigned to the random factor for rb designs.

Num_exp sets the number of independent measurements per group.

For one-way designs, the user provides Group_means as a vector. Number of levels are recognised based on number of means. For two-way designs, two vectors are to be provided by the user containing means of levels of a second factor. Number of means in both vectors should be the same. These functions can only handle balanced designs, i.e. same number of observations in all groups.

The output is a data frame with one or two columns denoting the fixed factor with levels that match the number of means entered. For rb data, the column for RandFac denotes levels of the blocking factor. The quantitative response variables are in the numeric Values column.

Value

This function produces a data.frame object containing simulated data.

Examples

#Basic usage with two levels within FactorX2, 
#20 experiments with inter-experiment SD 20, and residual SD 15

two_rb_tab <- make_2way_rb_data(c(100, 20), c(200, 300), 20, 20, 15)

str(two_rb_tab)
head(two_rb_tab)

Description

The make_1way_data, make_1way_rb_data, make_2way_data and make_2way_rb_data functions generate independent or randomised block (rb) design data of one-way or two-way designs.

Usage

make_2way_data(Group_1_means, Group_2_means, Num_obs, Residual_SD)

Arguments

Details

Random variates from the normal distribution based on user provided mean and SD provided are generated. For independent designs, the Residual_SD argument is used to set expected residual SD from the linear model. Exp_SD is used to set experiment-to-experiment SD, that will be assigned to the random factor for rb designs.

Num_obs sets the number of independent measurements per group.

For one-way designs, the user provides Group_means as a vector. Number of levels are recognised based on number of means. For two-way designs, two vectors are to be provided by the user containing means of levels of a second factor. Number of means in both vectors should be the same. These functions can only handle balanced designs, i.e. same number of observations in all groups.

The output is a data frame with one or two columns denoting the fixed factor with levels that match the number of means entered. For rb data, the column for RandFac denotes levels of the blocking factor. The quantitative response variables are in the numeric Values column.

Value

This function produces a data.frame object containing simulated data.

Examples

#Basic usage with two levels within FactorX2, 20 observations in each group, with residual SD 15

two_independent_tab <- make_2way_data(c(100, 20), c(200, 300), 20, 15)

#Four levels with 5 observations and residual SD 5
two_independent_tab <- make_2way_data(c(100, 20, 1500, 20), c(150, 5, 1450, 25), 5, 5)

Description

The make_1way_data, make_1way_rb_data, make_2way_data and make_2way_rb_data functions generate independent or randomised block (rb) design data of one-way or two-way designs.

Usage

make_2way_rb_data(Group1_means, Group2_means, Num_exp, Exp_SD, Residual_SD)

Arguments

Details

Random variates from the normal distribution based on user provided mean and SD provided are generated. For independent designs, the Residual_SD argument is used to set expected residual SD from the linear model. Exp_SD is used to set experiment-to-experiment SD, that will be assigned to the random factor (RandFac) for rb designs.

Num_exp sets the number of independent measurements per group.

For one-way designs, the user provides Group_means as a vector. Number of levels are recognised based on number of means. For two-way designs, two vectors are to be provided by the user containing means of levels of a second factor. Number of means in both vectors should be the same. These functions can only handle balanced designs, i.e. same number of observations in all groups.

The output is a data frame with one or two columns denoting the fixed factor with levels that match the number of means entered. For rb data, the column for RandFac denotes levels of the blocking factor. The quantitative response variables are in the numeric Values column.

Value

This function produces a data.frame object containing simulated data.

Examples

#Basic usage with two levels within FactorX2, 
#20 experiments with inter-experiment SD 20, and residual SD 15

two_rb_tab <- make_2way_rb_data(c(100, 20), c(200, 300), 20, 20, 15)

str(two_rb_tab)
head(two_rb_tab)

Description

One of four related functions for mixed effects analyses (based on lmer and as_lmerModLmerTest) to get a linear model for downstream steps, or an ANOVA table.

1. mixed_model

2. mixed_anova

3. mixed_model_slopes

4. mixed_anova_slopes.

Usage

mixed_anova(
  data,
  Y_value,
  Fixed_Factor,
  Random_Factor,
  Df_method = "Kenward-Roger",
  SS_method = "II",
  ...
)

Arguments

Details

These functions require a data table, one dependent variable (Y_value), one or more independent variables (Fixed_Factor), and at least one random factor (Random_Factor). These should match names of variables in the long-format data table exactly.

Outputs of mixed_model and mixed_model_slopes can be used for post-hoc comparisons with posthoc_Pairwise, posthoc_Levelwise, posthoc_vsRef, posthoc_Trends_Pairwise, posthoc_Trends_Levelwise and posthoc_Trends_vsRefor with emmeans.

More than one fixed factors can be provided as a vector (e.g. c("A", "B")). A full model with interaction term is fitted. This means when Y_value = Y, Fixed_factor = c("A", "B"), Random_factor = "R" are entered as arguments, these are passed on as Y ~ A*B + (1|R) (which is equivalent to Y ~ A + B + A:B + (1|R)).

In mixed_model_slopes and mixed_anova_slopes, the following kind of formula is used: Y ~ A*B + (S|R) (which is equivalent to Y ~ A + B + A:B + (S|R)). In this experimental implementation, random slopes and intercepts are fitted ((Slopes_Factor|Random_Factor)). Only one term each is allowed for Slopes_Factor and Random_Factor.

Value

ANOVA table of class "anova" and "data.frame".

Examples

#Usage with one fixed (Student) and random factor (Experiment)
mixed_anova(data = data_doubling_time, 
Y_value = "Doubling_time", 
Fixed_Factor = "Student", 
Random_Factor = "Experiment")

#two fixed factors provided as a vector
mixed_anova(data = data_cholesterol, 
Y_value = "Cholesterol", 
Fixed_Factor = c("Treatment", "Hospital"), 
Random_Factor = "Subject")

Description

One of four related functions for mixed effects analyses (based on lmer and as_lmerModLmerTest) to get a linear model for downstream steps, or an ANOVA table.

1. mixed_model

2. mixed_anova

3. mixed_model_slopes

4. mixed_anova_slopes.

Usage

mixed_anova_slopes(
  data,
  Y_value,
  Fixed_Factor,
  Slopes_Factor,
  Random_Factor,
  Df_method = "Kenward-Roger",
  SS_method = "II",
  ...
)

Arguments

Details

These functions require a data table, one dependent variable (Y_value), one or more independent variables (Fixed_Factor), and at least one random factor (Random_Factor). These should match names of variables in the long-format data table exactly.

Outputs of mixed_model and mixed_model_slopes can be used for post-hoc comparisons with posthoc_Pairwise, posthoc_Levelwise, posthoc_vsRef, posthoc_Trends_Pairwise, posthoc_Trends_Levelwise and posthoc_Trends_vsRefor with emmeans.

More than one fixed factors can be provided as a vector (e.g. c("A", "B")). A full model with interaction term is fitted. This means when Y_value = Y, Fixed_factor = c("A", "B"), Random_factor = "R" are entered as arguments, these are passed on as Y ~ A*B + (1|R) (which is equivalent to Y ~ A + B + A:B + (1|R)).

In mixed_model_slopes and mixed_anova_slopes, the following kind of formula is used: Y ~ A*B + (S|R) (which is equivalent to Y ~ A + B + A:B + (S|R)). In this experimental implementation, random slopes and intercepts are fitted ((Slopes_Factor|Random_Factor)). Only one term each is allowed for Slopes_Factor and Random_Factor.

Value

ANOVA table of class "anova" and "data.frame".

Examples

mixed_anova_slopes(data = data_2w_Tdeath,
Y_value = "PI",
Fixed_Factor = c("Genotype", "Time"),
Slopes_Factor = "Time",
Random_Factor = "Experiment")

Description

One of four related functions for mixed effects analyses (based on lmer and as_lmerModLmerTest) to get a linear model for downstream steps, or an ANOVA table.

1. mixed_model

2. mixed_anova

3. mixed_model_slopes

4. mixed_anova_slopes.

Usage

mixed_model(data, Y_value, Fixed_Factor, Random_Factor, ...)

Arguments

Details

These functions require a data table, one dependent variable (Y_value), one or more independent variables (Fixed_Factor), and at least one random factor (Random_Factor). These should match names of variables in the long-format data table exactly.

Outputs of mixed_model and mixed_model_slopes can be used for post-hoc comparisons with posthoc_Pairwise, posthoc_Levelwise, posthoc_vsRef, posthoc_Trends_Pairwise, posthoc_Trends_Levelwise and posthoc_Trends_vsRefor with emmeans.

More than one fixed factors can be provided as a vector (e.g. c("A", "B")). A full model with interaction term is fitted. This means when Y_value = Y, Fixed_factor = c("A", "B"), Random_factor = "R" are entered as arguments, these are passed on as Y ~ A*B + (1|R) (which is equivalent to Y ~ A + B + A:B + (1|R)).

In mixed_model_slopes and mixed_anova_slopes, the following kind of formula is used: Y ~ A*B + (S|R) (which is equivalent to Y ~ A + B + A:B + (S|R)). In this experimental implementation, random slopes and intercepts are fitted ((Slopes_Factor|Random_Factor)). Only one term each is allowed for Slopes_Factor and Random_Factor.

Value

This function returns an S4 object of class "lmerModLmerTest".

Examples

#one fixed factor and random factor
mixed_model(data = data_doubling_time, 
Y_value = "Doubling_time", 
Fixed_Factor = "Student", 
Random_Factor = "Experiment")

#two fixed factors as a vector, one random factor
mixed_model(data = data_cholesterol, 
Y_value = "Cholesterol", 
Fixed_Factor = c("Treatment", "Hospital"), 
Random_Factor = "Subject")

#save model
model <- mixed_model(data = data_doubling_time, 
Y_value =  "Doubling_time", 
Fixed_Factor = "Student", 
Random_Factor = "Experiment")

#get model summary
summary(model)

Description

One of four related functions for mixed effects analyses (based on lmer and as_lmerModLmerTest) to get a linear model for downstream steps, or an ANOVA table.

1. mixed_model

2. mixed_anova

3. mixed_model_slopes

4. mixed_anova_slopes.

Usage

mixed_model_slopes(
  data,
  Y_value,
  Fixed_Factor,
  Slopes_Factor,
  Random_Factor,
  ...
)

Arguments

Details

These functions require a data table, one dependent variable (Y_value), one or more independent variables (Fixed_Factor), and at least one random factor (Random_Factor). These should match names of variables in the long-format data table exactly.

Outputs of mixed_model and mixed_model_slopes can be used for post-hoc comparisons with posthoc_Pairwise, posthoc_Levelwise, posthoc_vsRef, posthoc_Trends_Pairwise, posthoc_Trends_Levelwise and posthoc_Trends_vsRefor with emmeans.

More than one fixed factors can be provided as a vector (e.g. c("A", "B")). A full model with interaction term is fitted. This means when Y_value = Y, Fixed_factor = c("A", "B"), Random_factor = "R" are entered as arguments, these are passed on as Y ~ A*B + (1|R) (which is equivalent to Y ~ A + B + A:B + (1|R)).

In mixed_model_slopes and mixed_anova_slopes, the following kind of formula is used: Y ~ A*B + (S|R) (which is equivalent to Y ~ A + B + A:B + (S|R)). In this experimental implementation, random slopes and intercepts are fitted ((Slopes_Factor|Random_Factor)). Only one term each is allowed for Slopes_Factor and Random_Factor.

Value

This function returns an S4 object of class "lmerModLmerTest".

Examples

#two fixed factors as a vector, 
#exactly one slope factor and random factor
mod <- mixed_model_slopes(data = data_2w_Tdeath,
Y_value = "PI",
Fixed_Factor = c("Genotype", "Time"),
Slopes_Factor = "Time",
Random_Factor = "Experiment")
#get summary
summary(mod)

Description

One of 4 related functions for plotting 1-way ANOVA designs with a blocking factor.

1. plot_3d_point_sd (mean & SD, SEM or CI95 error bars)

2. plot_3d_scatterbar (bar & SD, SEM or CI95 error bars)

3. plot_3d_scatterbox (box & whiskers)

4. plot_3d_scatterviolin (box & whiskers, violin)

Usage

plot_3d_point_sd(
  data,
  xcol,
  ycol,
  shapes,
  facet,
  ErrorType = "SD",
  symsize = 3.5,
  s_alpha = 1,
  symshape = 22,
  all_alpha = 0.3,
  all_size = 2.5,
  all_shape = 0,
  all_jitter = 0,
  ewid = 0.2,
  TextXAngle = 0,
  LogYTrans,
  LogYBreaks = waiver(),
  LogYLabels = waiver(),
  LogYLimits = NULL,
  facet_scales = "fixed",
  fontsize = 20,
  symthick,
  ethick,
  ColPal = c("okabe_ito", "all_grafify", "bright", "contrast", "dark", "fishy", "kelly",
    "light", "muted", "pale", "r4", "safe", "vibrant"),
  ColSeq = TRUE,
  ColRev = FALSE,
  SingleColour = "NULL",
  ...
)

Arguments

Details

The blocking factor (or any other categorical variable) can be mapped to the shapes argument (up to 25 levels allowed). Variables passed to xcol and shapes are internally converted to factors even if they are numeric or other type of variables.

In plot_3d_point_sd and plot_3d_scatterbar, the default error bar is SD (can be changed to SEM or CI95). In plot_3d_point_sd, a large coloured symbol is plotted at the mean, all other data are shown as smaller symbols. Boxplot uses geom_boxplot to depict median (thicker line), box (interquartile range (IQR)) and the whiskers (1.5*IQR).

Colours can be changed using ColPal, ColRev or ColSeq arguments. ColPal can be one of the following: "okabe_ito", "dark", "light", "bright", "pale", "vibrant, "muted" or "contrast". ColRev (logical TRUE/FALSE) decides whether colours are chosen from first-to-last or last-to-first from within the chosen palette. ColSeq (logical TRUE/FALSE) decides whether colours are picked by respecting the order in the palette or the most distant ones using colorRampPalette.

The resulting ggplot2 graph can take additional geometries or other layers.

Value

This function returns a ggplot2 object of class "gg" and "ggplot".

Examples

#3d version for 1-way data with blocking
#use plot_point_sd when no a blocking factor is not used
plot_3d_point_sd(data = data_1w_death, 
xcol = Genotype, ycol = Death, 
shapes = Experiment)

Description

One of 4 related functions for plotting 1-way ANOVA designs with a blocking factor.

1. plot_3d_point_sd (mean & SD, SEM or CI95 error bars)

2. plot_3d_scatterbar (bar & SD, SEM or CI95 error bars)

3. plot_3d_scatterbox (box & whiskers)

4. plot_3d_scatterviolin (box & whiskers, violin)

Usage

plot_3d_scatterbar(
  data,
  xcol,
  ycol,
  shapes,
  facet,
  ErrorType = "SD",
  symsize = 3,
  s_alpha = 0.8,
  b_alpha = 1,
  jitter = 0.1,
  ewid = 0.2,
  TextXAngle = 0,
  LogYTrans,
  LogYBreaks = waiver(),
  LogYLabels = waiver(),
  LogYLimits = NULL,
  facet_scales = "fixed",
  fontsize = 20,
  symthick,
  bthick,
  ColPal = c("okabe_ito", "all_grafify", "bright", "contrast", "dark", "fishy", "kelly",
    "light", "muted", "pale", "r4", "safe", "vibrant"),
  ColSeq = TRUE,
  ColRev = FALSE,
  SingleColour = "NULL",
  ...
)

Arguments

Details

The blocking factor (or any other categorical variable) can be mapped to the shapes argument (up to 25 levels allowed). Variables passed to xcol and shapes are internally converted to factors even if they are numeric or other type of variables.

In plot_3d_point_sd and plot_3d_scatterbar, the default error bar is SD (can be changed to SEM or CI95). In plot_3d_point_sd, a large coloured symbol is plotted at the mean, all other data are shown as smaller symbols. Boxplot uses geom_boxplot to depict median (thicker line), box (interquartile range (IQR)) and the whiskers (1.5*IQR).

Colours can be changed using ColPal, ColRev or ColSeq arguments. ColPal can be one of the following: "okabe_ito", "dark", "light", "bright", "pale", "vibrant, "muted" or "contrast". ColRev (logical TRUE/FALSE) decides whether colours are chosen from first-to-last or last-to-first from within the chosen palette. ColSeq (logical TRUE/FALSE) decides whether colours are picked by respecting the order in the palette or the most distant ones using colorRampPalette.

The resulting ggplot2 graph can take additional geometries or other layers.

Value

This function returns a ggplot2 object of class "gg" and "ggplot".

Examples

#3d version for 1-way data with blocking
#use plot_scatterbar_sd without blocking factor
plot_3d_scatterbar(data = data_1w_death, 
xcol = Genotype, ycol = Death, 
shapes = Experiment)

Description

One of 4 related functions for plotting 1-way ANOVA designs with a blocking factor.

1. plot_3d_point_sd (mean & SD, SEM or CI95 error bars)

2. plot_3d_scatterbar (bar & SD, SEM or CI95 error bars)

3. plot_3d_scatterbox (box & whiskers)

4. plot_3d_scatterviolin (box & whiskers, violin)

Usage

plot_3d_scatterbox(
  data,
  xcol,
  ycol,
  shapes,
  facet,
  symsize = 3,
  s_alpha = 0.8,
  b_alpha = 1,
  bwid = 0.5,
  jitter = 0.1,
  TextXAngle = 0,
  LogYTrans,
  LogYBreaks = waiver(),
  LogYLabels = waiver(),
  LogYLimits = NULL,
  facet_scales = "fixed",
  fontsize = 20,
  symthick,
  bthick,
  ColPal = c("okabe_ito", "all_grafify", "bright", "contrast", "dark", "fishy", "kelly",
    "light", "muted", "pale", "r4", "safe", "vibrant"),
  ColSeq = TRUE,
  ColRev = FALSE,
  SingleColour = "NULL",
  ...
)

Arguments

Details

The blocking factor (or any other categorical variable) can be mapped to the shapes argument (up to 25 levels allowed). Variables passed to xcol and shapes are internally converted to factors even if they are numeric or other type of variables.

In plot_3d_point_sd and plot_3d_scatterbar, the default error bar is SD (can be changed to SEM or CI95). In plot_3d_point_sd, a large coloured symbol is plotted at the mean, all other data are shown as smaller symbols. Boxplot uses geom_boxplot to depict median (thicker line), box (interquartile range (IQR)) and the whiskers (1.5*IQR).

Colours can be changed using ColPal, ColRev or ColSeq arguments. ColPal can be one of the following: "okabe_ito", "dark", "light", "bright", "pale", "vibrant, "muted" or "contrast". ColRev (logical TRUE/FALSE) decides whether colours are chosen from first-to-last or last-to-first from within the chosen palette. ColSeq (logical TRUE/FALSE) decides whether colours are picked by respecting the order in the palette or the most distant ones using colorRampPalette.

The resulting ggplot2 graph can take additional geometries or other layers.

Value

This function returns a ggplot2 object of class "gg" and "ggplot".

Examples

#3d version for 1-way data with blocking
plot_3d_scatterbox(data = data_1w_death, 
xcol = Genotype, ycol = Death, 
shapes = Experiment)
#use plot_scatterbox without a blocking factor

Description

One of 4 related functions for plotting 1-way ANOVA designs with a blocking factor.

1. plot_3d_point_sd (mean & SD, SEM or CI95 error bars)

2. plot_3d_scatterbar (bar & SD, SEM or CI95 error bars)

3. plot_3d_scatterbox (box & whiskers)

4. plot_3d_scatterviolin (box & whiskers, violin)

Usage

plot_3d_scatterviolin(
  data,
  xcol,
  ycol,
  shapes,
  facet,
  symsize = 3,
  s_alpha = 0.8,
  b_alpha = 0,
  v_alpha = 1,
  bwid = 0.3,
  vadjust = 1,
  jitter = 0.1,
  TextXAngle = 0,
  scale = "width",
  trim = TRUE,
  LogYTrans,
  LogYBreaks = waiver(),
  LogYLabels = waiver(),
  LogYLimits = NULL,
  facet_scales = "fixed",
  fontsize = 20,
  symthick,
  bthick,
  vthick,
  bvthick,
  ColPal = c("okabe_ito", "all_grafify", "bright", "contrast", "dark", "fishy", "kelly",
    "light", "muted", "pale", "r4", "safe", "vibrant"),
  ColSeq = TRUE,
  ColRev = FALSE,
  SingleColour = "NULL",
  ...
)

Arguments

Details

The blocking factor (or any other categorical variable) can be mapped to the shapes argument (up to 25 levels allowed). Variables passed to xcol and shapes are internally converted to factors even if they are numeric or other type of variables.

In plot_3d_point_sd and plot_3d_scatterbar, the default error bar is SD (can be changed to SEM or CI95). In plot_3d_point_sd, a large coloured symbol is plotted at the mean, all other data are shown as smaller symbols. Boxplot uses geom_boxplot to depict median (thicker line), box (interquartile range (IQR)) and the whiskers (1.5*IQR).

Colours can be changed using ColPal, ColRev or ColSeq arguments. ColPal can be one of the following: "okabe_ito", "dark", "light", "bright", "pale", "vibrant, "muted" or "contrast". ColRev (logical TRUE/FALSE) decides whether colours are chosen from first-to-last or last-to-first from within the chosen palette. ColSeq (logical TRUE/FALSE) decides whether colours are picked by respecting the order in the palette or the most distant ones using colorRampPalette.

The resulting ggplot2 graph can take additional geometries or other layers.

Value

This function returns a ggplot2 object of class "gg" and "ggplot".

Examples

#3d version for 1-way data with blocking
plot_3d_scatterviolin(data = data_1w_death, 
xcol = Genotype, ycol = Death, 
shapes = Experiment)
#use plot_scatterviolin without a blocking factor

Description

There are 4 related functions for 2-way ANOVA type plots. In addition to a categorical variable along the X-axis, a grouping factor is passed to either points, bars or boxes argument in these functions. A blocking factor (or any other categorical variable) can be optionally passed to the shapes argument.

1. plot_4d_point_sd (mean & SD, SEM or CI95 error bars)

2. plot_4d_scatterbar (bar & SD, SEM or CI95 error bars)

3. plot_4d_scatterbox (box & whiskers)

4. plot_4d_scatterviolin (box & whiskers, violin)

Usage

plot_4d_point_sd(
  data,
  xcol,
  ycol,
  points,
  shapes,
  facet,
  ErrorType = "SD",
  symsize = 3.5,
  s_alpha = 1,
  symshape = 22,
  all_alpha = 0.3,
  all_size = 2.5,
  all_shape = 0,
  all_jitter = 0,
  ewid = 0.2,
  group_wid = 0.8,
  TextXAngle = 0,
  LogYTrans,
  LogYBreaks = waiver(),
  LogYLabels = waiver(),
  LogYLimits = NULL,
  facet_scales = "fixed",
  fontsize = 20,
  symthick,
  ethick,
  ColPal = c("okabe_ito", "all_grafify", "bright", "contrast", "dark", "fishy", "kelly",
    "light", "muted", "pale", "r4", "safe", "vibrant"),
  ColSeq = TRUE,
  ColRev = FALSE,
  ...
)

Arguments

Details

These can be especially useful when the fourth variable shapes is a random factor or blocking factor (up to 25 levels are allowed; there will be an error with more levels). The shapes argument can be left blank to plot ordinary 2-way ANOVAs without blocking.

In plot_4d_point_sd and plot_4d_scatterbar, the default error bar is SD (can be changed to SEM or CI95). In plot_4d_point_sd, a large coloured symbol is plotted at the mean, all other data are shown as smaller symbols. Boxplot uses geom_boxplot to depict median (thicker line), box (interquartile range (IQR)) and the whiskers (1.5*IQR).

Colours can be changed using ColPal, ColRev or ColSeq arguments. ColPal can be one of the following: "okabe_ito", "dark", "light", "bright", "pale", "vibrant, "muted" or "contrast". ColRev (logical TRUE/FALSE) decides whether colours are chosen from first-to-last or last-to-first from within the chosen palette. ColSeq (logical TRUE/FALSE) decides whether colours are picked by respecting the order in the palette or the most distant ones using colorRampPalette.

The resulting ggplot2 graph can take additional geometries or other layers.

Value

This function returns a ggplot2 object of class "gg" and "ggplot".

Examples

#4d version for 2-way data with blocking
plot_4d_point_sd(data = data_2w_Tdeath, 
xcol = Genotype, 
ycol = PI, 
points = Time, 
shapes = Experiment)

#4d version without blocking factor
#`shapes` can be left blank
plot_4d_point_sd(data = data_2w_Festing, 
xcol = Strain, 
ycol = GST, 
points = Treatment)

Description

There are 4 related functions for 2-way ANOVA type plots. In addition to a categorical variable along the X-axis, a grouping factor is passed to either points, bars or boxes argument in these functions. A blocking factor (or any other categorical variable) can be optionally passed to the shapes argument.

1. plot_4d_point_sd (mean & SD, SEM or CI95 error bars)

2. plot_4d_scatterbar (bar & SD, SEM or CI95 error bars)

3. plot_4d_scatterbox (box & whiskers)

4. plot_4d_scatterviolin (box & whiskers, violin)

Usage

plot_4d_scatterbar(
  data,
  xcol,
  ycol,
  bars,
  shapes,
  facet,
  ErrorType = "SD",
  symsize = 3,
  s_alpha = 0.8,
  b_alpha = 1,
  bwid = 0.7,
  jitter = 0.1,
  ewid = 0.2,
  TextXAngle = 0,
  LogYTrans,
  LogYBreaks = waiver(),
  LogYLabels = waiver(),
  LogYLimits = NULL,
  facet_scales = "fixed",
  fontsize = 20,
  group_wid = 0.8,
  symthick,
  bthick,
  ColPal = c("okabe_ito", "all_grafify", "bright", "contrast", "dark", "fishy", "kelly",
    "light", "muted", "pale", "r4", "safe", "vibrant"),
  ColRev = FALSE,
  ColSeq = TRUE,
  ...
)

Arguments

Details

These can be especially useful when the fourth variable shapes is a random factor or blocking factor (up to 25 levels are allowed; there will be an error with more levels). The shapes argument can be left blank to plot ordinary 2-way ANOVAs without blocking.

In plot_4d_point_sd and plot_4d_scatterbar, the default error bar is SD (can be changed to SEM or CI95). In plot_4d_point_sd, a large coloured symbol is plotted at the mean, all other data are shown as smaller symbols. Boxplot uses geom_boxplot to depict median (thicker line), box (interquartile range (IQR)) and the whiskers (1.5*IQR).

Colours can be changed using ColPal, ColRev or ColSeq arguments. ColPal can be one of the following: "okabe_ito", "dark", "light", "bright", "pale", "vibrant, "muted" or "contrast". ColRev (logical TRUE/FALSE) decides whether colours are chosen from first-to-last or last-to-first from within the chosen palette. ColSeq (logical TRUE/FALSE) decides whether colours are picked by respecting the order in the palette or the most distant ones using colorRampPalette.

The resulting ggplot2 graph can take additional geometries or other layers.

Value

This function returns a ggplot2 object of class "gg" and "ggplot".

Examples

#4d version for 2-way data with blocking
plot_4d_scatterbar(data = data_2w_Tdeath, 
xcol = Genotype, 
ycol = PI, 
bars = Time, 
shapes = Experiment)

#4d version without blocking factor
#`shapes` can be left blank
plot_4d_scatterbar(data = data_2w_Tdeath, 
xcol = Genotype, 
ycol = PI, 
bars = Time)

Description

There are 4 related functions for 2-way ANOVA type plots. In addition to a categorical variable along the X-axis, a grouping factor is passed to either points, bars or boxes argument in these functions. A blocking factor (or any other categorical variable) can be optionally passed to the shapes argument.

1. plot_4d_point_sd (mean & SD, SEM or CI95 error bars)

2. plot_4d_scatterbar (bar & SD, SEM or CI95 error bars)

3. plot_4d_scatterbox (box & whiskers)

4. plot_4d_scatterviolin (box & whiskers, violin)

Usage

plot_4d_scatterbox(
  data,
  xcol,
  ycol,
  boxes,
  shapes,
  facet,
  symsize = 3,
  s_alpha = 0.8,
  b_alpha = 1,
  bwid = 0.7,
  jitter = 0.1,
  TextXAngle = 0,
  LogYTrans,
  LogYBreaks = waiver(),
  LogYLabels = waiver(),
  LogYLimits = NULL,
  facet_scales = "fixed",
  fontsize = 20,
  group_wid = 0.8,
  symthick,
  bthick,
  ColPal = c("okabe_ito", "all_grafify", "bright", "contrast", "dark", "fishy", "kelly",
    "light", "muted", "pale", "r4", "safe", "vibrant"),
  ColSeq = TRUE,
  ColRev = FALSE,
  ...
)

Arguments

Details

These can be especially useful when the fourth variable shapes is a random factor or blocking factor (up to 25 levels are allowed; there will be an error with more levels). The shapes argument can be left blank to plot ordinary 2-way ANOVAs without blocking.

In plot_4d_point_sd and plot_4d_scatterbar, the default error bar is SD (can be changed to SEM or CI95). In plot_4d_point_sd, a large coloured symbol is plotted at the mean, all other data are shown as smaller symbols. Boxplot uses geom_boxplot to depict median (thicker line), box (interquartile range (IQR)) and the whiskers (1.5*IQR).

Colours can be changed using ColPal, ColRev or ColSeq arguments. ColPal can be one of the following: "okabe_ito", "dark", "light", "bright", "pale", "vibrant, "muted" or "contrast". ColRev (logical TRUE/FALSE) decides whether colours are chosen from first-to-last or last-to-first from within the chosen palette. ColSeq (logical TRUE/FALSE) decides whether colours are picked by respecting the order in the palette or the most distant ones using colorRampPalette.

The resulting ggplot2 graph can take additional geometries or other layers.

Value

This function returns a ggplot2 object of class "gg" and "ggplot".

Examples

#4d version for 2-way data with blocking
plot_4d_scatterbox(data = data_2w_Tdeath, 
xcol = Genotype, 
ycol = PI, 
boxes = Time, 
shapes = Experiment)

#4d version without blocking factor
#`shapes` can be left blank
plot_4d_scatterbox(data = data_2w_Tdeath, 
xcol = Genotype, 
ycol = PI, 
boxes = Time)

Description

There are 4 related functions for 2-way ANOVA type plots. In addition to a categorical variable along the X-axis, a grouping factor is passed to either points, bars or boxes argument in these functions. A blocking factor (or any other categorical variable) can be optionally passed to the shapes argument.

1. plot_4d_point_sd (mean & SD, SEM or CI95 error bars)

2. plot_4d_scatterbar (bar & SD, SEM or CI95 error bars)

3. plot_4d_scatterbox (box & whiskers)

4. plot_4d_scatterviolin (box & whiskers, violin)

Usage

plot_4d_scatterviolin(
  data,
  xcol,
  ycol,
  boxes,
  shapes,
  facet,
  symsize = 3,
  s_alpha = 0.8,
  v_alpha = 1,
  b_alpha = 0,
  bwid = 0.3,
  vadjust = 1,
  jitter = 0.1,
  TextXAngle = 0,
  scale = "width",
  trim = TRUE,
  LogYTrans,
  LogYBreaks = waiver(),
  LogYLabels = waiver(),
  LogYLimits = NULL,
  facet_scales = "fixed",
  fontsize = 20,
  group_wid = 0.8,
  symthick,
  bthick,
  vthick,
  bvthick,
  ColPal = c("okabe_ito", "all_grafify", "bright", "contrast", "dark", "fishy", "kelly",
    "light", "muted", "pale", "r4", "safe", "vibrant"),
  ColSeq = TRUE,
  ColRev = FALSE,
  ...
)

Arguments

Details

These can be especially useful when the fourth variable shapes is a random factor or blocking factor (up to 25 levels are allowed; there will be an error with more levels). The shapes argument can be left blank to plot ordinary 2-way ANOVAs without blocking.

In plot_4d_point_sd and plot_4d_scatterbar, the default error bar is SD (can be changed to SEM or CI95). In plot_4d_point_sd, a large coloured symbol is plotted at the mean, all other data are shown as smaller symbols. Boxplot uses geom_boxplot to depict median (thicker line), box (interquartile range (IQR)) and the whiskers (1.5*IQR).

Colours can be changed using ColPal, ColRev or ColSeq arguments. ColPal can be one of the following: "okabe_ito", "dark", "light", "bright", "pale", "vibrant, "muted" or "contrast". ColRev (logical TRUE/FALSE) decides whether colours are chosen from first-to-last or last-to-first from within the chosen palette. ColSeq (logical TRUE/FALSE) decides whether colours are picked by respecting the order in the palette or the most distant ones using colorRampPalette.

The resulting ggplot2 graph can take additional geometries or other layers.

Value

This function returns a ggplot2 object of class "gg" and "ggplot".

Examples

#4d version for 2-way data with blocking
plot_4d_scatterviolin(data = data_2w_Tdeath, 
xcol = Genotype, 
ycol = PI, 
boxes = Time, 
shapes = Experiment)

#4d version without blocking factor
#`shapes` can be left blank
plot_4d_scatterviolin(data = data_2w_Tdeath, 
xcol = Genotype, 
ycol = PI, 
boxes = Time)

Description

One of 3 related functions to plot matching data joined by lines. The variable containing information for matching (e.g. matched subjects or experiments etc.) is passed to the match argument.

1. plot_befafter_colours or plot_befafter_colors,

2. plot_befafter_shapes

3. plot_befafter_box

Usage

plot_befafter_box(
  data,
  xcol,
  ycol,
  match,
  facet,
  PlotShapes = FALSE,
  symsize = 3,
  s_alpha = 0.8,
  b_alpha = 1,
  bwid = 0.4,
  jitter = 0.1,
  TextXAngle = 0,
  LogYTrans,
  LogYBreaks = waiver(),
  LogYLabels = waiver(),
  LogYLimits = NULL,
  facet_scales = "fixed",
  fontsize = 20,
  symthick,
  bthick,
  lthick,
  ColPal = c("okabe_ito", "all_grafify", "bright", "contrast", "dark", "fishy", "kelly",
    "light", "muted", "pale", "r4", "safe", "vibrant"),
  ColSeq = TRUE,
  ColRev = FALSE,
  SingleColour = "NULL",
  ...
)

Arguments

Details

In plot_befafter_colours/plot_befafter_colors and plot_befafter_shapes setting Boxplot = TRUE will also plot a box and whiskers plot.

Note that only 25 shapes are available, and there will be errors with plot_befafter_shapes when there are fewer than 25 matched observations; instead use plot_befafter_colours instead.

Colours can be changed using ColPal, ColRev or ColSeq arguments. ColPal can be one of the following: "okabe_ito", "dark", "light", "bright", "pale", "vibrant, "muted" or "contrast". ColRev (logical TRUE/FALSE) decides whether colours are chosen from first-to-last or last-to-first from within the chosen palette. ColSeq decides whether colours are picked by respecting the order in the palette or the most distant ones using colorRampPalette.

To plot a graph with a single colour along the X axis variable, use the SingleColour argument.

The resulting ggplot2 graph can take additional geometries or other layers.

Value

This function returns a ggplot2 object of class "gg" and "ggplot".

Examples

plot_befafter_box(data = data_t_pdiff, 
xcol = Condition, ycol = Mass, 
match = Subject)

#with PlotShapes = TRUE
plot_befafter_box(data = data_t_pdiff, 
xcol = Condition, ycol = Mass, 
match = Subject, PlotShapes = TRUE)

Description

One of 3 related functions to plot matching data joined by lines. The variable containing information for matching (e.g. matched subjects or experiments etc.) is passed to the match argument.

1. plot_befafter_colours or plot_befafter_colors,

2. plot_befafter_shapes

3. plot_befafter_box

Usage

plot_befafter_colours(
  data,
  xcol,
  ycol,
  match,
  facet,
  Boxplot = FALSE,
  symsize = 3,
  s_alpha = 1,
  jitter = 0.1,
  bwid = 0.4,
  TextXAngle = 0,
  LogYTrans,
  LogYBreaks = waiver(),
  LogYLabels = waiver(),
  LogYLimits = NULL,
  facet_scales = "fixed",
  fontsize = 20,
  symthick,
  bthick,
  lthick,
  ColPal = c("okabe_ito", "all_grafify", "bright", "contrast", "dark", "fishy", "kelly",
    "light", "muted", "pale", "r4", "safe", "vibrant"),
  ColSeq = TRUE,
  ColRev = FALSE,
  SingleColour = "NULL",
  ...
)

plot_befafter_colors(
  data,
  xcol,
  ycol,
  match,
  facet,
  Boxplot = FALSE,
  symsize = 3,
  s_alpha = 1,
  jitter = 0.1,
  bwid = 0.4,
  TextXAngle = 0,
  LogYTrans,
  LogYBreaks = waiver(),
  LogYLabels = waiver(),
  LogYLimits = NULL,
  facet_scales = "fixed",
  fontsize = 20,
  symthick,
  bthick,
  lthick,
  ColPal = c("okabe_ito", "all_grafify", "bright", "contrast", "dark", "fishy", "kelly",
    "light", "muted", "pale", "r4", "safe", "vibrant"),
  ColSeq = TRUE,
  ColRev = FALSE,
  SingleColour = "NULL",
  ...
)

Arguments

Details

In plot_befafter_colours/plot_befafter_colors and plot_befafter_shapes setting Boxplot = TRUE will also plot a box and whiskers plot.

Note that only 25 shapes are available, and there will be errors with plot_befafter_shapes when there are fewer than 25 matched observations; instead use plot_befafter_colours instead.

Colours can be changed using ColPal, ColRev or ColSeq arguments. ColPal can be one of the following: "okabe_ito", "dark", "light", "bright", "pale", "vibrant, "muted" or "contrast". ColRev (logical TRUE/FALSE) decides whether colours are chosen from first-to-last or last-to-first from within the chosen palette. ColSeq decides whether colours are picked by respecting the order in the palette or the most distant ones using colorRampPalette.

To plot a graph with a single colour along the X axis variable, use the SingleColour argument.

The resulting ggplot2 graph can take additional geometries or other layers.

Value

This function returns a ggplot2 object of class "gg" and "ggplot".

Examples

#plot without legends if necessary
plot_befafter_colours(data = data_t_pdiff, 
xcol = Condition, ycol = Mass, 
match = Subject, s_alpha = .9, ColSeq = FALSE)+
guides(fill = "none", 
colour = "none") #remove guides
#plot with boxplot
plot_befafter_colours(data = data_t_pdiff, 
xcol = Condition, ycol = Mass, 
match = Subject, s_alpha = .9, ColSeq = FALSE,
Boxplot = TRUE)+
guides(fill = "none", 
colour = "none") #remove guides

Description

One of 3 related functions to plot matching data joined by lines. The variable containing information for matching (e.g. matched subjects or experiments etc.) is passed to the match argument.

1. plot_befafter_colours or plot_befafter_colors,

2. plot_befafter_shapes

3. plot_befafter_box

Usage

plot_befafter_shapes(
  data,
  xcol,
  ycol,
  match,
  facet,
  Boxplot = FALSE,
  symsize = 3,
  s_alpha = 1,
  bwid = 0.4,
  jitter = 0.1,
  TextXAngle = 0,
  LogYTrans,
  LogYBreaks = waiver(),
  LogYLabels = waiver(),
  LogYLimits = NULL,
  facet_scales = "fixed",
  fontsize = 20,
  symthick,
  bthick,
  lthick,
  ColPal = c("okabe_ito", "all_grafify", "bright", "contrast", "dark", "fishy", "kelly",
    "light", "muted", "pale", "r4", "safe", "vibrant"),
  ColSeq = TRUE,
  ColRev = FALSE,
  SingleColour = "NULL",
  ...
)

Arguments

Details

In plot_befafter_colours/plot_befafter_colors and plot_befafter_shapes setting Boxplot = TRUE will also plot a box and whiskers plot.

Note that only 25 shapes are available, and there will be errors with plot_befafter_shapes when there are fewer than 25 matched observations; instead use plot_befafter_colours instead.

Colours can be changed using ColPal, ColRev or ColSeq arguments. ColPal can be one of the following: "okabe_ito", "dark", "light", "bright", "pale", "vibrant, "muted" or "contrast". ColRev (logical TRUE/FALSE) decides whether colours are chosen from first-to-last or last-to-first from within the chosen palette. ColSeq decides whether colours are picked by respecting the order in the palette or the most distant ones using colorRampPalette.

To plot a graph with a single colour along the X axis variable, use the SingleColour argument.

The resulting ggplot2 graph can take additional geometries or other layers.

Value

This function returns a ggplot2 object of class "gg" and "ggplot".

Examples

#plot without legends if necessary
plot_befafter_colors(data = data_t_pdiff, 
xcol = Condition, ycol = Mass, 
match = Subject, s_alpha = .9, ColSeq = FALSE)+
guides(fill = "none", 
colour = "none") #remove guides
#plot with boxplot
plot_befafter_colors(data = data_t_pdiff, 
xcol = Condition, ycol = Mass, 
match = Subject, s_alpha = .9, ColSeq = FALSE,
Boxplot = TRUE)+
guides(fill = "none", 
colour = "none") #remove guides

Description

This function takes a data table, ycol of quantitative variable and a categorical grouping variable (group), if available, and plots a density graph using geom_density). Alternatives are plot_histogram, or plot_qqline.

Usage

plot_density(
  data,
  ycol,
  group,
  facet,
  PlotType = c("Density", "Counts", "Normalised counts"),
  c_alpha = 0.2,
  TextXAngle = 0,
  facet_scales = "fixed",
  fontsize = 20,
  linethick,
  LogYTrans,
  LogYBreaks = waiver(),
  LogYLabels = waiver(),
  LogYLimits = NULL,
  ColPal = c("okabe_ito", "all_grafify", "bright", "contrast", "dark", "fishy", "kelly",
    "light", "muted", "pale", "r4", "safe", "vibrant"),
  ColSeq = TRUE,
  ColRev = FALSE,
  SingleColour = NULL,
  ...
)

Arguments

Details

Note that the function requires the quantitative Y variable first, and groups them based on an X variable. The group variable is mapped to the fill and colour aesthetics in geom_density. Colours can be changed using ColPal, ColRev or ColSeq arguments. Colours available can be seen quickly with plot_grafify_palette. ColPal can be one of the following: "okabe_ito", "dark", "light", "bright", "pale", "vibrant, "muted" or "contrast". ColRev (logical TRUE/FALSE) decides whether colours are chosen from first-to-last or last-to-first from within the chosen palette. ColSeq decides whether colours are picked by respecting the order in the palette or the most distant ones using colorRampPalette.

Value

This function returns a ggplot2 object of class "gg" and "ggplot".

Examples

plot_density(data = data_t_pratio, 
ycol = log(Cytokine), group = Genotype)

#with faceting
plot_density(data = data_cholesterol, 
ycol = Cholesterol, group = Treatment, 
facet = Treatment, fontsize = 12)

#Counts
plot_density(data = data_cholesterol, 
ycol = Cholesterol, group = Treatment,
PlotType = "Counts", 
facet = Treatment, fontsize = 12)

Description

There are three types of plot_dot_ functions that plot data as "dots" using the geom_dotplot geometry. They all take a data table, a categorical X variable and a numeric Y variable.

1. plot_dotbar_sd (bar & SD, SEM or CI95 error bars)

2. plot_dotbox (box & whiskers)

3. plot_dotviolin (box & whiskers, violin)

Usage

plot_dotbar_sd(
  data,
  xcol,
  ycol,
  facet,
  ErrorType = "SD",
  dotsize = 1.5,
  d_alpha = 0.8,
  b_alpha = 1,
  bwid = 0.5,
  ewid = 0.2,
  TextXAngle = 0,
  LogYTrans,
  LogYBreaks = waiver(),
  LogYLabels = waiver(),
  LogYLimits = NULL,
  facet_scales = "fixed",
  fontsize = 20,
  dotthick,
  bthick,
  ColPal = c("okabe_ito", "all_grafify", "bright", "contrast", "dark", "fishy", "kelly",
    "light", "muted", "pale", "r4", "safe", "vibrant"),
  ColSeq = TRUE,
  ColRev = FALSE,
  SingleColour = "NULL",
  ...
)

Arguments

Details

Related plot_scatter_ variants show data symbols using the geom_point geometry. These are plot_scatterbar_sd (or SEM or CI95 error bars), plot_scatterbox and plot_scatterviolin. Over plotting in plot_scatter variants can be reduced with the jitter argument.

The X variable is mapped to the fill aesthetic of dots, symbols, bars, boxes and violins.

Colours can be changed using ColPal, ColRev or ColSeq arguments. Colours available can be seen quickly with plot_grafify_palette. ColPal can be one of the following: "okabe_ito", "dark", "light", "bright", "pale", "vibrant, "muted" or "contrast". ColRev (logical TRUE/FALSE) decides whether colours are chosen from first-to-last or last-to-first from within the chosen palette. ColSeq decides whether colours are picked by respecting the order in the palette or the most distant ones using colorRampPalette.

If you prefer a single colour for the graph, use the SingleColour argument.

Value

This function returns a ggplot2 object of class "gg" and "ggplot".

Examples

plot_dotbar_sd(data = data_cholesterol, 
xcol = Treatment,
ycol = Cholesterol)

plot_dotbar_sd(data = data_1w_death, 
xcol = Genotype, ycol = Death, 
ColPal = "pale", ColSeq = FALSE, ColRev = TRUE)

#single colour along X
plot_dotbar_sd(data = data_1w_death, 
xcol = Genotype, ycol = Death, 
SingleColour = "light_orange")

Description

There are three types of plot_dot_ functions that plot data as "dots" using the geom_dotplot geometry. They all take a data table, a categorical X variable and a numeric Y variable.

1. plot_dotbar_sd (bar & SD, SEM or CI95 error bars)

2. plot_dotbox (box & whiskers)

3. plot_dotviolin (box & whiskers, violin)

Usage

plot_dotbox(
  data,
  xcol,
  ycol,
  facet,
  dotsize = 1.5,
  d_alpha = 0.8,
  b_alpha = 1,
  bwid = 0.5,
  TextXAngle = 0,
  LogYTrans,
  LogYBreaks = waiver(),
  LogYLabels = waiver(),
  LogYLimits = NULL,
  facet_scales = "fixed",
  fontsize = 20,
  dotthick,
  bthick,
  ColPal = c("okabe_ito", "all_grafify", "bright", "contrast", "dark", "fishy", "kelly",
    "light", "muted", "pale", "r4", "safe", "vibrant"),
  ColSeq = TRUE,
  ColRev = FALSE,
  SingleColour = "NULL",
  ...
)

Arguments

Details

Related plot_scatter_ variants show data symbols using the geom_point geometry. These are plot_scatterbar_sd (or SEM or CI95 error bars), plot_scatterbox and plot_scatterviolin. Over plotting in plot_scatter variants can be reduced with the jitter argument.

The X variable is mapped to the fill aesthetic of dots, symbols, bars, boxes and violins.

Colours can be changed using ColPal, ColRev or ColSeq arguments. Colours available can be seen quickly with plot_grafify_palette. ColPal can be one of the following: "okabe_ito", "dark", "light", "bright", "pale", "vibrant, "muted" or "contrast". ColRev (logical TRUE/FALSE) decides whether colours are chosen from first-to-last or last-to-first from within the chosen palette. ColSeq decides whether colours are picked by respecting the order in the palette or the most distant ones using colorRampPalette.

If you prefer a single colour for the graph, use the SingleColour argument.

Value

This function returns a ggplot2 object of class "gg" and "ggplot".

Examples

plot_dotbox(data = data_1w_death, 
xcol = Genotype, ycol = Death)

plot_dotbox(data = data_1w_death, 
xcol = Genotype, ycol = Death, 
ColPal = "vibrant", b_alpha = 0.5)
plot_dotbox(data = data_1w_death, 
xcol = Genotype, ycol = Death, 
SingleColour = "safe_bluegreen", b_alpha = 0.5)

Description

There are three types of plot_dot_ functions that plot data as "dots" using the geom_dotplot geometry. They all take a data table, a categorical X variable and a numeric Y variable.

1. plot_dotbar_sd (bar & SD, SEM or CI95 error bars)

2. plot_dotbox (box & whiskers)

3. plot_dotviolin (box & whiskers, violin)

Usage

plot_dotviolin(
  data,
  xcol,
  ycol,
  facet,
  dotsize = 1.5,
  d_alpha = 0.8,
  b_alpha = 0,
  v_alpha = 1,
  bwid = 0.3,
  vadjust = 1,
  trim = TRUE,
  scale = "width",
  TextXAngle = 0,
  LogYTrans,
  LogYBreaks = waiver(),
  LogYLabels = waiver(),
  LogYLimits = NULL,
  facet_scales = "fixed",
  fontsize = 20,
  dotthick,
  bthick,
  vthick,
  bvthick,
  ColPal = c("okabe_ito", "all_grafify", "bright", "contrast", "dark", "fishy", "kelly",
    "light", "muted", "pale", "r4", "safe", "vibrant"),
  ColSeq = TRUE,
  ColRev = FALSE,
  SingleColour = "NULL",
  ...
)

Arguments

Details

Related plot_scatter_ variants show data symbols using the geom_point geometry. These are plot_scatterbar_sd (or SEM or CI95 error bars), plot_scatterbox and plot_scatterviolin. Over plotting in plot_scatter variants can be reduced with the jitter argument.

The X variable is mapped to the fill aesthetic of dots, symbols, bars, boxes and violins.

Colours can be changed using ColPal, ColRev or ColSeq arguments. Colours available can be seen quickly with plot_grafify_palette. ColPal can be one of the following: "okabe_ito", "dark", "light", "bright", "pale", "vibrant, "muted" or "contrast". ColRev (logical TRUE/FALSE) decides whether colours are chosen from first-to-last or last-to-first from within the chosen palette. ColSeq decides whether colours are picked by respecting the order in the palette or the most distant ones using colorRampPalette.

If you prefer a single colour for the graph, use the SingleColour argument.

Value

This function returns a ggplot2 object of class "gg" and "ggplot".

Examples

#plot with trim = FALSE
plot_dotviolin(data = data_t_pdiff, 
xcol = Condition, ycol = Mass, 
dotsize = 2, trim = FALSE)

plot_dotviolin(data = data_t_pdiff, 
xcol = Condition, ycol = Mass,
trim = FALSE, b_alpha = 0.5, 
ColPal = "pale", ColSeq = FALSE)

#single colour along X
plot_dotviolin(data = data_t_pdiff, 
xcol = Condition, ycol = Mass,
trim = FALSE, b_alpha = 0.5, 
SingleColour = "pale_cyan")

Description

Plot prediction of gam model

Usage

plot_gam_predict(
  Model,
  xcol,
  ycol,
  ByFactor,
  symsize = 1,
  s_alpha = 0.1,
  smooth_alpha = 0.7,
  linethick,
  fontsize = 20,
  ...
)

Arguments

Value

This function returns a ggplot2 object of class "gg" and "ggplot".

Examples

#fit zooplankton data
z1 <- ga_model(data = data_zooplankton,
Y_value = "log(density_adj)",
Fixed_Factor = "taxon",
Smooth_Factor = "day")

#plot fitted data
plot_gam_predict(Model = z1,
xcol = day, 
ycol = `log(density_adj)`,
ByFactor = taxon)

Description

This simple function allows quick visualisation of colours in grafify palettes and their hex codes. It uses plot_scatterbar_sd and some arguments are similar and can be adjusted.

Usage

plot_grafify_palette(
  palette = c("okabe_ito", "all_grafify", "bright", "contrast", "dark", "fishy", "kelly",
    "light", "muted", "pale", "r4", "safe", "vibrant", "OrBl_div", "PrGn_div",
    "blue_conti", "grey_conti", "yellow_conti"),
  fontsize = 14,
  ...
)

Arguments

Value

This function returns a ggplot2 object of class "gg" and "ggplot".

Examples

plot_grafify_palette("pale")
plot_grafify_palette("contrast")

Description

This function takes a data table, a quantitative variable (ycol) and a grouping variable (group), if available, and plots a histogram graph using geom_histogram). Alternatives are plot_density, or plot_qqline.

Usage

plot_histogram(
  data,
  ycol,
  group,
  facet,
  PlotType = c("Counts", "Normalised counts"),
  BinSize = 30,
  c_alpha = 0.8,
  TextXAngle = 0,
  facet_scales = "fixed",
  fontsize = 20,
  linethick,
  alpha,
  LogYTrans,
  LogYBreaks = waiver(),
  LogYLabels = waiver(),
  LogYLimits = NULL,
  ColPal = c("okabe_ito", "all_grafify", "bright", "contrast", "dark", "fishy", "kelly",
    "light", "muted", "pale", "r4", "safe", "vibrant"),
  ColSeq = TRUE,
  ColRev = FALSE,
  SingleColour = NULL,
  ...
)

Arguments

Details

Note that the function requires the quantitative Y variable first, and groups them based on a categorical variable passed on via the group argument. The grouping variable is mapped to the fill aesthetics in geom_histogram.

ColPal & ColRev options are applied to both fill and colour scales. Colours available can be seen quickly with plot_grafify_palette. Colours can be changed using ColPal, ColRev or ColSeq arguments. ColPal can be one of the following: "okabe_ito", "dark", "light", "bright", "pale", "vibrant, "muted" or "contrast". ColRev (logical TRUE/FALSE) decides whether colours are chosen from first-to-last or last-to-first from within the chosen palette. ColSeq decides whether colours are picked by respecting the order in the palette or the most distant ones using colorRampPalette.

Value

This function returns a ggplot2 object of class "gg" and "ggplot".

Examples

#Basic usage
plot_histogram(data = data_t_pratio, 
ycol = Cytokine, group = Genotype, 
BinSize = 10)
#with log transformation
plot_histogram(data = data_t_pratio, 
ycol = log(Cytokine), group = Genotype, 
BinSize = 10)
#Normalised counts
plot_histogram(data = data_t_pratio, 
ycol = log(Cytokine), group = Genotype, 
PlotType = "Normalised counts", 
BinSize = 10)

Description

This function takes a linear model, and up to three variables and plots observe data (circles) and model predictions (squares). If the X-variable is categorical, a box and whiskers plot is overlaid. A variable (ByFactor) can be used for faceting.

Usage

plot_lm_predict(
  Model,
  xcol,
  ycol,
  ByFactor,
  obs_size = 2,
  obs_alpha = 0.3,
  pred_size = 2,
  pred_alpha = 0.8,
  linethick,
  base_size = 15,
  ...
)

Arguments

Value

This function returns a ggplot2 object of class "gg" and "ggplot".

Examples

#fit a model
deathm1 <- mixed_model(data_2w_Tdeath, 
"PI", c("Genotype", "Time"), 
"Experiment")
#plot model
plot_lm_predict(deathm1,
Genotype, PI, Time)
#fit zooplankton data
z1 <- ga_model(data = data_zooplankton,
Y_value = "log(density_adj)",
Fixed_Factor = "taxon",
Smooth_Factor = "day")

#plot fitted data
plot_lm_predict(Model = z1,
xcol = day, 
ycol = `log(density_adj)`,
ByFactor = taxon)

Description

This function allows "log10" or "log2" transformation of X or Y axes. With "log10" transformation, log10 ticks are also added on the outside.

Usage

plot_logscale(
  Plot,
  LogYTrans = "log10",
  LogXTrans,
  LogYBreaks = waiver(),
  LogXBreaks = waiver(),
  LogYLimits = NULL,
  LogXLimits = NULL,
  LogYLabels = waiver(),
  LogXLabels = waiver(),
  fontsize = 22,
  ...
)

Arguments

Details

Arguments allow for axes limits, breaks and labels to passed on.

Value

This function returns a ggplot2 object of class "gg" and "ggplot".

Examples

#save a ggplot object
P <- ggplot(data_t_pratio, 
aes(Genotype,Cytokine))+
geom_jitter(shape = 21, 
size = 5, width = .2, 
aes(fill = Genotype), 
alpha = .7)
#transform Y axis
plot_logscale(Plot = P)

#or in one go 
plot_logscale(ggplot(data_t_pratio, 
aes(Genotype,Cytokine))+
geom_jitter(shape = 21, 
size = 5, width = .2, 
aes(fill = Genotype), 
alpha = .7))

Description

There are 4 related functions that use geom_point to plot a categorical variable along the X axis.

1. plot_point_sd (mean & SD, SEM or CI95 error bars)

2. plot_scatterbar_sd (bar & SD, SEM or CI95 error bars)

3. plot_scatterbox (box & whiskers)

4. plot_scatterviolin (box & whiskers, violin)

Usage

plot_point_sd(
  data,
  xcol,
  ycol,
  facet,
  ErrorType = "SD",
  symsize = 3.5,
  s_alpha = 1,
  symshape = 22,
  all_alpha = 0.3,
  all_size = 2.5,
  all_shape = 1,
  all_jitter = 0,
  ewid = 0.2,
  TextXAngle = 0,
  LogYTrans,
  LogYBreaks = waiver(),
  LogYLabels = waiver(),
  LogYLimits = NULL,
  facet_scales = "fixed",
  fontsize = 20,
  symthick,
  ethick,
  ColPal = c("okabe_ito", "all_grafify", "bright", "contrast", "dark", "fishy", "kelly",
    "light", "muted", "pale", "r4", "safe", "vibrant"),
  ColSeq = TRUE,
  ColRev = FALSE,
  SingleColour = "NULL",
  ...
)

Arguments