bryan/hartman-server
1
0
Fork 0
Code Issues Pull Requests Releases Wiki Activity

Refactor data loading and add more plotting elements

Browse Source
This commit is contained in:
Bryan Roessler
2024-09-17 14:35:39 -04:00
parent a58c8db90d
commit f342e292c8
1 changed files with 256 additions and 201 deletions
Download Patch File Download Diff File Expand all files Collapse all files

457
qhtcp-workflow/apps/r/calculate_interaction_zscores.R
View File

@@ -37,28 +37,30 @@ parse_arguments <- function() {
} else { } else {
commandArgs(trailingOnly = TRUE) commandArgs(trailingOnly = TRUE)
} }
# Extract paths, names, and standard deviations out_dir <- normalizePath(args[1], mustWork = FALSE)
paths <- args[seq(4, length(args), by = 3)] sgd_gene_list <- normalizePath(args[2], mustWork = FALSE)
names <- args[seq(5, length(args), by = 3)] easy_results_file <- normalizePath(args[3], mustWork = FALSE)
sds <- as.numeric(args[seq(6, length(args), by = 3)])
# The remaining arguments should be in groups of 3
# Normalize paths exp_args <- args[-(1:3)]
normalized_paths <- normalizePath(paths, mustWork = FALSE) if (length(exp_args) %% 3 != 0) {
stop("Experiment arguments should be in groups of 3: path, name, sd.")
# Create named list of experiments }
experiments <- list() experiments <- list()
for (i in seq_along(paths)) { for (i in seq(1, length(exp_args), by = 3)) {
experiments[[names[i]]] <- list( exp_name <- exp_args[i + 1]
path = normalized_paths[i], experiments[[exp_name]] <- list(
sd = sds[i] path = normalizePath(exp_args[i], mustWork = FALSE),
sd = as.numeric(exp_args[i + 2])
) )
} }
list( list(
out_dir = normalizePath(args[1], mustWork = FALSE), out_dir = out_dir,
sgd_gene_list = normalizePath(args[2], mustWork = FALSE), sgd_gene_list = sgd_gene_list,
easy_results_file = normalizePath(args[3], mustWork = FALSE), easy_results_file = easy_results_file,
experiments = experiments experiments = experiments
) )
} }
@@ -81,9 +83,11 @@ theme_publication <- function(base_size = 14, base_family = "sans", legend_posit
plot.background = element_rect(colour = NA), plot.background = element_rect(colour = NA),
panel.border = element_rect(colour = NA), panel.border = element_rect(colour = NA),
axis.title = element_text(face = "bold", size = rel(1)), axis.title = element_text(face = "bold", size = rel(1)),
axis.title.y = element_text(angle = 90, vjust = 2), axis.title.y = element_text(angle = 90, vjust = 2, size = 18),
axis.title.x = element_text(vjust = -0.2), axis.title.x = element_text(vjust = -0.2, size = 18),
axis.line = element_line(colour = "black"), axis.line = element_line(colour = "black"),
axis.text.x = element_text(size = 16),
axis.text.y = element_text(size = 16),
panel.grid.major = element_line(colour = "#f0f0f0"), panel.grid.major = element_line(colour = "#f0f0f0"),
panel.grid.minor = element_blank(), panel.grid.minor = element_blank(),
legend.key = element_rect(colour = NA), legend.key = element_rect(colour = NA),
@@ -111,27 +115,42 @@ scale_colour_publication <- function(...) {
# Load the initial dataframe from the easy_results_file # Load the initial dataframe from the easy_results_file
load_and_process_data <- function(easy_results_file, sd = 3) { load_and_process_data <- function(easy_results_file, sd = 3) {
df <- read.delim(easy_results_file, skip = 2, as.is = TRUE, row.names = 1, strip.white = TRUE) df <- read.delim(
easy_results_file,
skip = 2,
stringsAsFactors = FALSE,
row.names = 1,
strip.white = TRUE
)
# Filter and rename columns
df <- df %>%
filter(!is.na(ORF) & ORF != "") %>%
filter(!Gene %in% c("BLANK", "Blank", "blank")) %>%
filter(Drug != "BMH21") %>%
rename(
L = l,
num = Num.,
AUC = AUC96,
scan = Scan,
last_bg = LstBackgrd,
first_bg = X1stBackgrd
) %>%
mutate(across(c(Col, Row, num, L, K, r, scan, AUC, last_bg, first_bg), as.numeric))
# Calculate delta background and tolerance
df <- df %>% df <- df %>%
filter(!(.[[1]] %in% c("", "Scan"))) %>%
filter(!is.na(ORF) & ORF != "" & !Gene %in% c("BLANK", "Blank", "blank") & Drug != "BMH21") %>%
# Rename columns
rename(L = l, num = Num., AUC = AUC96, scan = Scan, last_bg = LstBackgrd, first_bg = X1stBackgrd) %>%
mutate( mutate(
across(c(Col, Row, num, L, K, r, scan, AUC, last_bg, first_bg), as.numeric),
delta_bg = last_bg - first_bg, delta_bg = last_bg - first_bg,
delta_bg_tolerance = mean(delta_bg, na.rm = TRUE) + (sd * sd(delta_bg, na.rm = TRUE)), delta_bg_tolerance = mean(delta_bg, na.rm = TRUE) + (sd * sd(delta_bg, na.rm = TRUE)),
NG = if_else(L == 0 & !is.na(L), 1, 0), NG = if_else(L == 0 & !is.na(L), 1, 0),
DB = if_else(delta_bg >= delta_bg_tolerance, 1, 0), DB = if_else(delta_bg >= delta_bg_tolerance, 1, 0),
SM = 0, SM = 0,
OrfRep = if_else(ORF == "YDL227C", "YDL227C", OrfRep), # should these be hardcoded? OrfRep = if_else(ORF == "YDL227C", "YDL227C", OrfRep),
conc_num = as.numeric(gsub("[^0-9\\.]", "", Conc)), conc_num = as.numeric(gsub("[^0-9\\.]", "", Conc)),
conc_num_factor = as.factor(conc_num) conc_num_factor = as.factor(conc_num)
# conc_num_factor = factor(conc_num, levels = sort(unique(conc_num)))
# conc_num_numeric = as.numeric(conc_num_factor) - 1
) )
return(df) return(df)
} }
@@ -439,34 +458,43 @@ generate_and_save_plots <- function(output_dir, file_name, plot_configs, grid_la
dev.off() dev.off()
# Combine and save interactive HTML plots # Combine and save interactive HTML plots
combined_plot <- subplot(plotly_plots, nrows = grid_layout$nrow %||% length(plotly_plots), margin = 0.05) combined_plot <- subplot(plotly_plots,
nrows = ifelse(is.null(grid_layout$nrow), length(plotly_plots), grid_layout$nrow),
margin = 0.05)
saveWidget(combined_plot, file = file.path(output_dir, paste0(file_name, ".html")), selfcontained = TRUE) saveWidget(combined_plot, file = file.path(output_dir, paste0(file_name, ".html")), selfcontained = TRUE)
} }
generate_scatter_plot <- function(plot, config) { generate_scatter_plot <- function(plot, config) {
# 1. Determine Shape, Size, and Position for geom_point # Build the aes mapping with color if specified
if (!is.null(config$color_var)) {
plot <- plot + aes(color = .data[[config$color_var]])
}
# Determine Shape, Size, and Position for geom_point
shape <- if (!is.null(config$shape)) config$shape else 3 shape <- if (!is.null(config$shape)) config$shape else 3
size <- if (!is.null(config$size)) { size <- if (!is.null(config$size)) config$size else 0.1
config$size position <- if (!is.null(config$position) && config$position == "jitter") "jitter" else "identity"
} else {
if (!is.null(config$delta_bg_point) && config$delta_bg_point) 0.2 # Add geom_point with determined parameters
else if (!is.null(config$gene_point) && config$gene_point) 0.2 plot <- plot + geom_point(
else 0.1 aes(color = .data[[config$color_var]]),
shape = shape,
size = size,
position = position
)
if (!is.null(config$cyan_points)) {
plot <- plot + geom_point(
data = subset(config$df, is_cyan_point == TRUE),
aes(x = .data[[config$x_var]], y = .data[[config$y_var]]),
color = "cyan",
shape = 3,
size = 0.5
)
} }
position <- if (!is.null(config$delta_bg_point) && config$delta_bg_point) { # Add Smooth Line if specified
"identity"
} else if (!is.null(config$gene_point) && config$gene_point) {
"jitter"
} else {
if (!is.null(config$position) && config$position == "jitter") "jitter" else "identity"
}
# 2. Add geom_point with determined parameters
plot <- plot + geom_point(shape = shape, size = size, position = position)
# 3. Add Smooth Line if specified
if (!is.null(config$add_smooth) && config$add_smooth) { if (!is.null(config$add_smooth) && config$add_smooth) {
if (!is.null(config$lm_line)) { if (!is.null(config$lm_line)) {
plot <- plot + plot <- plot +
@@ -485,7 +513,7 @@ generate_scatter_plot <- function(plot, config) {
} }
} }
# 4. Add SD Bands if specified # Add SD Bands if specified
if (!is.null(config$sd_band_values)) { if (!is.null(config$sd_band_values)) {
for (sd_band in config$sd_band_values) { for (sd_band in config$sd_band_values) {
plot <- plot + plot <- plot +
@@ -509,8 +537,24 @@ generate_scatter_plot <- function(plot, config) {
) )
} }
} }
# Add Rectangles if specified
if (!is.null(config$rectangles)) {
for (rect in config$rectangles) {
plot <- plot + annotate(
"rect",
xmin = rect$xmin,
xmax = rect$xmax,
ymin = rect$ymin,
ymax = rect$ymax,
fill = ifelse(is.null(rect$fill), NA, rect$fill),
color = ifelse(is.null(rect$color), "black", rect$color),
alpha = ifelse(is.null(rect$alpha), 0.1, rect$alpha)
)
}
}
# 5. Add Error Bars if specified # Add Error Bars if specified
if (!is.null(config$error_bar) && config$error_bar && !is.null(config$y_var)) { if (!is.null(config$error_bar) && config$error_bar && !is.null(config$y_var)) {
y_mean_col <- paste0("mean_", config$y_var) y_mean_col <- paste0("mean_", config$y_var)
y_sd_col <- paste0("sd_", config$y_var) y_sd_col <- paste0("sd_", config$y_var)
@@ -525,7 +569,7 @@ generate_scatter_plot <- function(plot, config) {
) )
} }
# 6. Customize X-axis if specified # Customize X-axis if specified
if (!is.null(config$x_breaks) && !is.null(config$x_labels) && !is.null(config$x_label)) { if (!is.null(config$x_breaks) && !is.null(config$x_labels) && !is.null(config$x_label)) {
plot <- plot + plot <- plot +
scale_x_discrete( scale_x_discrete(
@@ -535,17 +579,17 @@ generate_scatter_plot <- function(plot, config) {
) )
} }
# 7. Apply coord_cartesian if specified # Apply coord_cartesian if specified
if (!is.null(config$coord_cartesian)) { if (!is.null(config$coord_cartesian)) {
plot <- plot + coord_cartesian(ylim = config$coord_cartesian) plot <- plot + coord_cartesian(ylim = config$coord_cartesian)
} }
# 8. Set Y-axis limits if specified # Set Y-axis limits if specified
if (!is.null(config$ylim_vals)) { if (!is.null(config$ylim_vals)) {
plot <- plot + scale_y_continuous(limits = config$ylim_vals) plot <- plot + scale_y_continuous(limits = config$ylim_vals)
} }
# 9. Add Annotations if specified # Add Annotations if specified
if (!is.null(config$annotations)) { if (!is.null(config$annotations)) {
for (annotation in config$annotations) { for (annotation in config$annotations) {
plot <- plot + plot <- plot +
@@ -559,12 +603,12 @@ generate_scatter_plot <- function(plot, config) {
} }
} }
# 10. Add Title if specified # Add Title if specified
if (!is.null(config$title)) { if (!is.null(config$title)) {
plot <- plot + ggtitle(config$title) plot <- plot + ggtitle(config$title)
} }
# 11. Adjust Legend Position if specified # Adjust Legend Position if specified
if (!is.null(config$legend_position)) { if (!is.null(config$legend_position)) {
plot <- plot + theme(legend.position = config$legend_position) plot <- plot + theme(legend.position = config$legend_position)
} }
@@ -692,121 +736,131 @@ generate_interaction_plot_configs <- function(df, variables) {
return(configs) return(configs)
} }
generate_rank_plot_configs <- function(df_filtered, variables, is_lm = FALSE, adjust = FALSE) { generate_rank_plot_configs <- function(df_filtered, variables, is_lm = FALSE) {
# Define SD bands
# Define SD bands sd_bands <- c(1, 2, 3)
sd_bands <- c(1, 2, 3)
# Initialize list to store plot configurations
# Initialize list to store plot configurations configs <- list()
configs <- list()
# SD-based plots for L and K
# SD-based plots for L and K for (variable in c("L", "K")) {
for (variable in c("L", "K")) { for (sd_band in sd_bands) {
for (sd_band in sd_bands) { # Determine columns based on whether it's lm or not
if (is_lm) {
# Determine columns based on whether it's lm or not rank_var <- paste0(variable, "_Rank_lm")
if (is_lm) { zscore_var <- paste0("Z_lm_", variable)
rank_var <- paste0(variable, "_Rank_lm") y_label <- paste("Int Z score", variable)
zscore_var <- paste0("Z_lm_", variable) } else {
y_label <- paste("Int Z score", variable) rank_var <- paste0(variable, "_Rank")
} else { zscore_var <- paste0("Avg_Zscore_", variable)
rank_var <- paste0(variable, "_Rank") y_label <- paste("Avg Z score", variable)
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)
# Annotated Plot Configuration num_suppressors <- sum(df_filtered[[zscore_var]] <= -sd_band, na.rm = TRUE)
configs[[length(configs) + 1]] <- list(
df = df_filtered, # Annotated Plot Configuration
x_var = rank_var, configs[[length(configs) + 1]] <- list(
y_var = zscore_var, df = df_filtered,
plot_type = "scatter", x_var = rank_var,
title = paste(y_label, "vs. Rank for", variable, "above", sd_band, "SD"), y_var = zscore_var,
sd_band = sd_band, plot_type = "scatter",
annotations = list( title = paste(y_label, "vs. Rank for", variable, "above", sd_band, "SD"),
list( sd_band = sd_band,
x = median(df_filtered[[rank_var]], na.rm = TRUE), annotations = list(
y = 10, list(
label = paste("Deletion Enhancers =", sum(df_filtered[[zscore_var]] >= sd_band, na.rm = TRUE)) x = median(df_filtered[[rank_var]], na.rm = TRUE),
), y = 10,
list( label = paste("Deletion Enhancers =", num_enhancers)
x = median(df_filtered[[rank_var]], na.rm = TRUE), ),
y = -10, list(
label = paste("Deletion Suppressors =", sum(df_filtered[[zscore_var]] <= -sd_band, na.rm = TRUE)) 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 sd_band_values = sd_band,
) shape = 3,
size = 0.1,
# Non-Annotated Plot Configuration y_label = y_label,
configs[[length(configs) + 1]] <- list( x_label = "Rank",
df = df_filtered, legend_position = "none"
x_var = rank_var, )
y_var = zscore_var,
plot_type = "scatter", # Non-Annotated Plot Configuration
title = paste(y_label, "vs. Rank for", variable, "above", sd_band, "SD No Annotations"), configs[[length(configs) + 1]] <- list(
sd_band = sd_band, df = df_filtered,
annotations = NULL, x_var = rank_var,
sd_band_values = sd_band, y_var = zscore_var,
shape = 3, plot_type = "scatter",
size = 0.1 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
# Avg ZScore and Rank Avg ZScore Plots for r, L, K, and AUC for (variable in variables) {
for (variable in variables) { for (plot_type in c("Avg_Zscore_vs_lm", "Rank_Avg_Zscore_vs_lm")) {
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") {
# Define x and y variables based on plot type x_var <- paste0("Avg_Zscore_", variable)
if (plot_type == "Avg_Zscore_vs_lm") { y_var <- paste0("Z_lm_", variable)
x_var <- paste0("Avg_Zscore_", variable) title_suffix <- paste("Avg Zscore vs lm", variable)
y_var <- paste0("Z_lm_", variable) } else {
title_suffix <- paste("Avg Zscore vs lm", variable) x_var <- paste0(variable, "_Rank")
} else if (plot_type == "Rank_Avg_Zscore_vs_lm") { y_var <- paste0(variable, "_Rank_lm")
x_var <- paste0(variable, "_Rank") title_suffix <- paste("Rank Avg Zscore vs lm", variable)
y_var <- paste0(variable, "_Rank_lm") }
title_suffix <- paste("Rank Avg Zscore vs lm", variable)
} # Fit linear model
lm_fit <- lm(df_filtered[[y_var]] ~ df_filtered[[x_var]], data = df_filtered)
# Determine y-axis label
if (plot_type == "Avg_Zscore_vs_lm") { # Check for perfect fit
y_label <- paste("Z lm", variable) if (summary(lm_fit)$sigma == 0) {
} else { next # Skip this iteration if the fit is perfect
y_label <- paste("Rank lm", variable) }
}
# Calculate R-squared
# Determine correlation text (R-squared) r_squared <- summary(lm_fit)$r.squared
lm_fit <- lm(df_filtered[[y_var]] ~ df_filtered[[x_var]], data = df_filtered)
r_squared <- summary(lm_fit)$r.squared # Plot Configuration
configs[[length(configs) + 1]] <- list(
# Plot Configuration df = df_filtered,
configs[[length(configs) + 1]] <- list( x_var = x_var,
df = df_filtered, y_var = y_var,
x_var = x_var, plot_type = "scatter",
y_var = y_var, title = title_suffix,
plot_type = "scatter", annotations = list(
title = title_suffix, list(
annotations = list( x = 0,
list( y = 0,
x = 0, label = paste("R-squared =", round(r_squared, 2))
y = 0, )
label = paste("R-squared =", round(r_squared, 2)) ),
) sd_band_values = NULL, # Not applicable
), shape = 3,
sd_band_values = NULL, # Not applicable size = 0.1,
shape = 3, add_smooth = TRUE,
size = 0.1, lm_line = list(intercept = coef(lm_fit)[1], slope = coef(lm_fit)[2]),
add_smooth = TRUE, legend_position = "right",
lm_line = list(intercept = coef(lm_fit)[1], slope = coef(lm_fit)[2]), color_var = "Overlap",
legend_position = "right" x_label = x_var,
) y_label = y_var
)
}
} }
}
return(configs)
return(configs)
} }
generate_correlation_plot_configs <- function(df) { generate_correlation_plot_configs <- function(df) {
@@ -831,7 +885,7 @@ generate_correlation_plot_configs <- function(df) {
config <- list( config <- list(
df = df, df = df,
x_var = rel$x, x_var = rel$x,
y_var = rel$y, y_var = rel.y,
plot_type = "scatter", plot_type = "scatter",
title = rel$label, title = rel$label,
x_label = paste("z-score", gsub("Z_lm_", "", rel$x)), x_label = paste("z-score", gsub("Z_lm_", "", rel$x)),
@@ -839,9 +893,19 @@ generate_correlation_plot_configs <- function(df) {
annotations = list( annotations = list(
list(x = 0, y = 0, label = paste("R-squared =", round(lm_summary$r.squared, 3))) list(x = 0, y = 0, label = paste("R-squared =", round(lm_summary$r.squared, 3)))
), ),
add_smooth = TRUE, # This flags that a geom_smooth layer should be added add_smooth = TRUE, # Add regression line
lm_line = list(intercept = coef(lm_model)[1], slope = coef(lm_model)[2]), # For direct geom_abline if needed lm_line = list(intercept = coef(lm_model)[1], slope = coef(lm_model)[2]),
legend_position = "right" legend_position = "right",
shape = 3,
size = 0.5,
color_var = "Overlap",
rectangles = list(
list(
xmin = -2, xmax = 2, ymin = -2, ymax = 2,
fill = NA, color = "grey20", alpha = 0.1
)
),
cyan_points = TRUE
) )
configs[[length(configs) + 1]] <- config configs[[length(configs) + 1]] <- config
@@ -850,6 +914,7 @@ generate_correlation_plot_configs <- function(df) {
return(configs) return(configs)
} }
filter_data <- function(df, variables, nf = FALSE, missing = FALSE, adjust = FALSE, filter_data <- function(df, variables, nf = FALSE, missing = FALSE, adjust = FALSE,
rank = FALSE, limits_map = NULL, verbose = TRUE) { rank = FALSE, limits_map = NULL, verbose = TRUE) {
@@ -1188,19 +1253,16 @@ main <- function() {
) )
) )
# message("Generating quality control 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, "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, "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, "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", 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_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", 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, "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, "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) generate_and_save_plots(out_dir_qc, "delta_background_vs_K_for_strains_2sd_outside_mean_K", delta_bg_outside_2sd_k_plots)
# TODO: Originally this filtered L NA's
# Let's try to avoid for now since stats have already been calculated
# Process background strains # Process background strains
bg_strains <- c("YDL227C") bg_strains <- c("YDL227C")
@@ -1349,8 +1411,7 @@ main <- function() {
rank_plot_configs <- generate_rank_plot_configs( rank_plot_configs <- generate_rank_plot_configs(
df = zscores_interactions_joined_filtered, df = zscores_interactions_joined_filtered,
variables = interaction_vars, variables = interaction_vars,
is_lm = FALSE, is_lm = FALSE
adjust = TRUE
) )
generate_and_save_plots(output_dir = out_dir, file_name = "RankPlots", generate_and_save_plots(output_dir = out_dir, file_name = "RankPlots",
plot_configs = rank_plot_configs, grid_layout = list(ncol = 3, nrow = 2)) plot_configs = rank_plot_configs, grid_layout = list(ncol = 3, nrow = 2))
@@ -1359,8 +1420,7 @@ main <- function() {
rank_lm_plot_configs <- generate_rank_plot_configs( rank_lm_plot_configs <- generate_rank_plot_configs(
df = zscores_interactions_joined_filtered, df = zscores_interactions_joined_filtered,
variables = interaction_vars, variables = interaction_vars,
is_lm = TRUE, is_lm = TRUE
adjust = TRUE
) )
generate_and_save_plots(output_dir = out_dir, file_name = "RankPlots_lm", generate_and_save_plots(output_dir = out_dir, file_name = "RankPlots_lm",
plot_configs = rank_lm_plot_configs, grid_layout = list(ncol = 3, nrow = 2)) plot_configs = rank_lm_plot_configs, grid_layout = list(ncol = 3, nrow = 2))
@@ -1370,8 +1430,6 @@ main <- function() {
zscores_interactions_filtered <- zscores_interactions_joined %>% zscores_interactions_filtered <- zscores_interactions_joined %>%
group_by(across(all_of(c("OrfRep", "Gene", "num")))) %>% group_by(across(all_of(c("OrfRep", "Gene", "num")))) %>%
filter(!is.na(Z_lm_L) | !is.na(Avg_Zscore_L)) %>% filter(!is.na(Z_lm_L) | !is.na(Avg_Zscore_L)) %>%
ungroup() %>%
rowwise() %>%
mutate( mutate(
lm_R_squared_L = if (n() > 1) summary(lm(Z_lm_L ~ Avg_Zscore_L))$r.squared else NA, lm_R_squared_L = if (n() > 1) summary(lm(Z_lm_L ~ Avg_Zscore_L))$r.squared else NA,
lm_R_squared_K = if (n() > 1) summary(lm(Z_lm_K ~ Avg_Zscore_K))$r.squared else NA, lm_R_squared_K = if (n() > 1) summary(lm(Z_lm_K ~ Avg_Zscore_K))$r.squared else NA,
@@ -1387,8 +1445,7 @@ main <- function() {
Z_lm_L <= -2 & Avg_Zscore_L >= 2 ~ "Deletion Suppressor lm, Deletion Enhancer Avg Z score", Z_lm_L <= -2 & Avg_Zscore_L >= 2 ~ "Deletion Suppressor lm, Deletion Enhancer Avg Z score",
TRUE ~ "No Effect" TRUE ~ "No Effect"
) )
) %>% )
ungroup()
# Re-rank # Re-rank
zscores_interactions_filtered <- filter_data( zscores_interactions_filtered <- filter_data(
@@ -1401,8 +1458,7 @@ main <- function() {
rank_plot_filtered_configs <- generate_rank_plot_configs( rank_plot_filtered_configs <- generate_rank_plot_configs(
df = zscores_interactions_filtered, df = zscores_interactions_filtered,
variables = interaction_vars, variables = interaction_vars,
is_lm = FALSE, is_lm = FALSE
adjust = FALSE
) )
message("Generating filtered ranked plots") message("Generating filtered ranked plots")
@@ -1416,8 +1472,7 @@ main <- function() {
rank_plot_lm_filtered_configs <- generate_rank_plot_configs( rank_plot_lm_filtered_configs <- generate_rank_plot_configs(
df = zscores_interactions_filtered, df = zscores_interactions_filtered,
variables = interaction_vars, variables = interaction_vars,
is_lm = TRUE, is_lm = TRUE
adjust = FALSE
) )
generate_and_save_plots( generate_and_save_plots(
output_dir = out_dir, output_dir = out_dir,