Reformat rank plot configs
This commit is contained in:
@@ -721,140 +721,132 @@ generate_interaction_plot_configs <- function(df, variables) {
|
||||
}
|
||||
|
||||
generate_rank_plot_configs <- function(df_filtered, variables, is_lm = FALSE) {
|
||||
# Define SD bands
|
||||
sd_bands <- c(1, 2, 3)
|
||||
|
||||
# Initialize list to store plot configurations
|
||||
configs <- list()
|
||||
|
||||
# SD-based plots for L and K
|
||||
for (variable in c("L", "K")) {
|
||||
for (sd_band in sd_bands) {
|
||||
# Determine columns based on whether it's lm or not
|
||||
if (is_lm) {
|
||||
rank_var <- paste0(variable, "_Rank_lm")
|
||||
zscore_var <- paste0("Z_lm_", variable)
|
||||
y_label <- paste("Int Z score", variable)
|
||||
} else {
|
||||
rank_var <- paste0(variable, "_Rank")
|
||||
zscore_var <- paste0("Avg_Zscore_", variable)
|
||||
y_label <- paste("Avg Z score", variable)
|
||||
}
|
||||
|
||||
# Calculate counts for annotations
|
||||
num_enhancers <- sum(df_filtered[[zscore_var]] >= sd_band, na.rm = TRUE)
|
||||
num_suppressors <- sum(df_filtered[[zscore_var]] <= -sd_band, na.rm = TRUE)
|
||||
|
||||
# Annotated Plot Configuration
|
||||
configs[[length(configs) + 1]] <- list(
|
||||
df = df_filtered,
|
||||
x_var = rank_var,
|
||||
y_var = zscore_var,
|
||||
plot_type = "scatter",
|
||||
title = paste(y_label, "vs. Rank for", variable, "above", sd_band, "SD"),
|
||||
sd_band = sd_band,
|
||||
annotations = list(
|
||||
list(
|
||||
x = median(df_filtered[[rank_var]], na.rm = TRUE),
|
||||
y = 10,
|
||||
label = paste("Deletion Enhancers =", num_enhancers)
|
||||
),
|
||||
list(
|
||||
x = median(df_filtered[[rank_var]], na.rm = TRUE),
|
||||
y = -10,
|
||||
label = paste("Deletion Suppressors =", num_suppressors)
|
||||
)
|
||||
),
|
||||
sd_band_values = sd_band,
|
||||
shape = 3,
|
||||
size = 0.1,
|
||||
y_label = y_label,
|
||||
x_label = "Rank",
|
||||
legend_position = "none"
|
||||
)
|
||||
|
||||
# Non-Annotated Plot Configuration
|
||||
configs[[length(configs) + 1]] <- list(
|
||||
df = df_filtered,
|
||||
x_var = rank_var,
|
||||
y_var = zscore_var,
|
||||
plot_type = "scatter",
|
||||
title = paste(y_label, "vs. Rank for", variable, "above", sd_band, "SD No Annotations"),
|
||||
sd_band = sd_band,
|
||||
annotations = NULL,
|
||||
sd_band_values = sd_band,
|
||||
shape = 3,
|
||||
size = 0.1,
|
||||
y_label = y_label,
|
||||
x_label = "Rank",
|
||||
legend_position = "none"
|
||||
)
|
||||
}
|
||||
|
||||
sd_bands <- c(1, 2, 3)
|
||||
|
||||
configs <- list()
|
||||
|
||||
# SD-based plots for L and K
|
||||
for (variable in c("L", "K")) {
|
||||
for (sd_band in sd_bands) {
|
||||
# Determine columns based on whether it's lm or not
|
||||
if (is_lm) {
|
||||
rank_var <- paste0(variable, "_Rank_lm")
|
||||
zscore_var <- paste0("Z_lm_", variable)
|
||||
y_label <- paste("Int Z score", variable)
|
||||
} else {
|
||||
rank_var <- paste0(variable, "_Rank")
|
||||
zscore_var <- paste0("Avg_Zscore_", variable)
|
||||
y_label <- paste("Avg Z score", variable)
|
||||
}
|
||||
|
||||
num_enhancers <- sum(df_filtered[[zscore_var]] >= sd_band, na.rm = TRUE)
|
||||
num_suppressors <- sum(df_filtered[[zscore_var]] <= -sd_band, na.rm = TRUE)
|
||||
|
||||
# Annotated plot configuration
|
||||
configs[[length(configs) + 1]] <- list(
|
||||
df = df_filtered,
|
||||
x_var = rank_var,
|
||||
y_var = zscore_var,
|
||||
plot_type = "scatter",
|
||||
title = paste(y_label, "vs. Rank for", variable, "above", sd_band, "SD"),
|
||||
sd_band = sd_band,
|
||||
annotations = list(
|
||||
list(
|
||||
x = median(df_filtered[[rank_var]], na.rm = TRUE),
|
||||
y = 10,
|
||||
label = paste("Deletion Enhancers =", num_enhancers)
|
||||
),
|
||||
list(
|
||||
x = median(df_filtered[[rank_var]], na.rm = TRUE),
|
||||
y = -10,
|
||||
label = paste("Deletion Suppressors =", num_suppressors)
|
||||
)
|
||||
),
|
||||
sd_band_values = sd_band,
|
||||
shape = 3,
|
||||
size = 0.1,
|
||||
y_label = y_label,
|
||||
x_label = "Rank",
|
||||
legend_position = "none"
|
||||
)
|
||||
|
||||
# Non-Annotated Plot Configuration
|
||||
configs[[length(configs) + 1]] <- list(
|
||||
df = df_filtered,
|
||||
x_var = rank_var,
|
||||
y_var = zscore_var,
|
||||
plot_type = "scatter",
|
||||
title = paste(y_label, "vs. Rank for", variable, "above", sd_band, "SD No Annotations"),
|
||||
sd_band = sd_band,
|
||||
annotations = NULL,
|
||||
sd_band_values = sd_band,
|
||||
shape = 3,
|
||||
size = 0.1,
|
||||
y_label = y_label,
|
||||
x_label = "Rank",
|
||||
legend_position = "none"
|
||||
)
|
||||
}
|
||||
}
|
||||
|
||||
# Avg ZScore and Rank Avg ZScore Plots for r, L, K, and AUC
|
||||
for (variable in variables) {
|
||||
for (plot_type in c("Avg_Zscore_vs_lm", "Rank_Avg_Zscore_vs_lm")) {
|
||||
# Define x and y variables based on plot type
|
||||
if (plot_type == "Avg_Zscore_vs_lm") {
|
||||
x_var <- paste0("Avg_Zscore_", variable)
|
||||
y_var <- paste0("Z_lm_", variable)
|
||||
title_suffix <- paste("Avg Zscore vs lm", variable)
|
||||
|
||||
# Add rectangles configuration for Avg_Zscore_vs_lm
|
||||
rectangles <- list(
|
||||
list(
|
||||
xmin = -2, xmax = 2, ymin = -2, ymax = 2,
|
||||
fill = NA, color = "grey20", alpha = 0.1
|
||||
)
|
||||
)
|
||||
} else {
|
||||
x_var <- paste0(variable, "_Rank")
|
||||
y_var <- paste0(variable, "_Rank_lm")
|
||||
title_suffix <- paste("Rank Avg Zscore vs lm", variable)
|
||||
rectangles <- NULL
|
||||
}
|
||||
|
||||
# Fit linear model
|
||||
lm_fit <- lm(df_filtered[[y_var]] ~ df_filtered[[x_var]], data = df_filtered)
|
||||
|
||||
# Check for perfect fit
|
||||
if (summary(lm_fit)$sigma == 0) {
|
||||
next # Skip this iteration if the fit is perfect
|
||||
}
|
||||
|
||||
# Calculate R-squared
|
||||
r_squared <- summary(lm_fit)$r.squared
|
||||
|
||||
# Plot Configuration
|
||||
configs[[length(configs) + 1]] <- list(
|
||||
df = df_filtered,
|
||||
x_var = x_var,
|
||||
y_var = y_var,
|
||||
plot_type = "scatter",
|
||||
title = title_suffix,
|
||||
annotations = list(
|
||||
list(
|
||||
x = 0,
|
||||
y = 0,
|
||||
label = paste("R-squared =", round(r_squared, 2))
|
||||
)
|
||||
),
|
||||
sd_band_values = NULL, # Not applicable
|
||||
shape = 3,
|
||||
size = 0.1,
|
||||
add_smooth = TRUE,
|
||||
lm_line = list(intercept = coef(lm_fit)[1], slope = coef(lm_fit)[2]),
|
||||
legend_position = "right",
|
||||
color_var = "Overlap",
|
||||
x_label = x_var,
|
||||
y_label = y_var,
|
||||
rectangles = rectangles # Add rectangles configuration
|
||||
)
|
||||
}
|
||||
# Avg ZScore and Rank Avg ZScore Plots for r, L, K, and AUC
|
||||
for (variable in variables) {
|
||||
for (plot_type in c("Avg_Zscore_vs_lm", "Rank_Avg_Zscore_vs_lm")) {
|
||||
# Define specific variables based on plot type
|
||||
if (plot_type == "Avg_Zscore_vs_lm") {
|
||||
x_var <- paste0("Avg_Zscore_", variable)
|
||||
y_var <- paste0("Z_lm_", variable)
|
||||
title_suffix <- paste("Avg Zscore vs lm", variable)
|
||||
rectangles <- list(
|
||||
list(xmin = -2, xmax = 2, ymin = -2, ymax = 2,
|
||||
fill = NA, color = "grey20", alpha = 0.1
|
||||
)
|
||||
)
|
||||
} else {
|
||||
x_var <- paste0(variable, "_Rank")
|
||||
y_var <- paste0(variable, "_Rank_lm")
|
||||
title_suffix <- paste("Rank Avg Zscore vs lm", variable)
|
||||
rectangles <- NULL
|
||||
}
|
||||
|
||||
# Fit linear model
|
||||
lm_fit <- lm(df_filtered[[y_var]] ~ df_filtered[[x_var]], data = df_filtered)
|
||||
|
||||
# Check for perfect fit
|
||||
if (summary(lm_fit)$sigma == 0) {
|
||||
next # Skip this iteration if the fit is perfect
|
||||
}
|
||||
|
||||
r_squared <- summary(lm_fit)$r.squared
|
||||
|
||||
configs[[length(configs) + 1]] <- list(
|
||||
df = df_filtered,
|
||||
x_var = x_var,
|
||||
y_var = y_var,
|
||||
plot_type = "scatter",
|
||||
title = title_suffix,
|
||||
annotations = list(
|
||||
list(
|
||||
x = 0,
|
||||
y = 0,
|
||||
label = paste("R-squared =", round(r_squared, 2))
|
||||
)
|
||||
),
|
||||
sd_band_values = NULL, # Not applicable
|
||||
shape = 3,
|
||||
size = 0.1,
|
||||
add_smooth = TRUE,
|
||||
lm_line = list(intercept = coef(lm_fit)[1], slope = coef(lm_fit)[2]),
|
||||
legend_position = "right",
|
||||
color_var = "Overlap",
|
||||
x_label = x_var,
|
||||
y_label = y_var,
|
||||
rectangles = rectangles # Add rectangles configuration
|
||||
)
|
||||
}
|
||||
|
||||
return(configs)
|
||||
}
|
||||
return(configs)
|
||||
}
|
||||
|
||||
generate_correlation_plot_configs <- function(df) {
|
||||
@@ -1247,16 +1239,16 @@ main <- function() {
|
||||
)
|
||||
)
|
||||
|
||||
message("Generating quality control plots")
|
||||
generate_and_save_plots(out_dir_qc, "L_vs_K_before_quality_control", l_vs_k_plots)
|
||||
generate_and_save_plots(out_dir_qc, "frequency_delta_background", frequency_delta_bg_plots)
|
||||
generate_and_save_plots(out_dir_qc, "L_vs_K_above_threshold", above_threshold_plots)
|
||||
generate_and_save_plots(out_dir_qc, "plate_analysis", plate_analysis_plots)
|
||||
generate_and_save_plots(out_dir_qc, "plate_analysis_boxplots", plate_analysis_boxplots)
|
||||
generate_and_save_plots(out_dir_qc, "plate_analysis_no_zeros", plate_analysis_no_zeros_plots)
|
||||
generate_and_save_plots(out_dir_qc, "plate_analysis_no_zeros_boxplots", plate_analysis_no_zeros_boxplots)
|
||||
generate_and_save_plots(out_dir_qc, "L_vs_K_for_strains_2SD_outside_mean_K", l_outside_2sd_k_plots)
|
||||
generate_and_save_plots(out_dir_qc, "delta_background_vs_K_for_strains_2sd_outside_mean_K", delta_bg_outside_2sd_k_plots)
|
||||
# message("Generating quality control plots")
|
||||
# generate_and_save_plots(out_dir_qc, "L_vs_K_before_quality_control", l_vs_k_plots)
|
||||
# generate_and_save_plots(out_dir_qc, "frequency_delta_background", frequency_delta_bg_plots)
|
||||
# generate_and_save_plots(out_dir_qc, "L_vs_K_above_threshold", above_threshold_plots)
|
||||
# generate_and_save_plots(out_dir_qc, "plate_analysis", plate_analysis_plots)
|
||||
# generate_and_save_plots(out_dir_qc, "plate_analysis_boxplots", plate_analysis_boxplots)
|
||||
# generate_and_save_plots(out_dir_qc, "plate_analysis_no_zeros", plate_analysis_no_zeros_plots)
|
||||
# generate_and_save_plots(out_dir_qc, "plate_analysis_no_zeros_boxplots", plate_analysis_no_zeros_boxplots)
|
||||
# generate_and_save_plots(out_dir_qc, "L_vs_K_for_strains_2SD_outside_mean_K", l_outside_2sd_k_plots)
|
||||
# generate_and_save_plots(out_dir_qc, "delta_background_vs_K_for_strains_2sd_outside_mean_K", delta_bg_outside_2sd_k_plots)
|
||||
|
||||
# Process background strains
|
||||
bg_strains <- c("YDL227C")
|
||||
|
||||
Reference in New Issue
Block a user