Upload New File

WES_plot_results.R

+library(dplyr)
+library(ggplot2)
+library(RColorBrewer)
+library(stringr)
+library(e1071)
+library(openxlsx)
+library(scales)
+library(reshape2)
+
+snv_colors2 <- function(del = FALSE, alpha = 1) {
+  nuc <- c("A", "C", "G", "T")
+  # Point deletions
+  dels <- alpha(brewer.pal(5, "Greys")[2:5], 1)
+  names(dels) <- paste0(nuc, "*")
+  # Transitions
+  ts <- alpha(brewer.pal(5, "Blues")[2:5], 1)
+  names(ts) <- c("CT", "GA", "AG", "TC")
+  # Tansversions
+  tv1 <- alpha(brewer.pal(5, "YlGn")[2:5], 1)
+  names(tv1) <- c("AC", "TG", "AT", "TA")
+  tv2 <- alpha(brewer.pal(5, "OrRd")[2:5], 1)
+  names(tv2) <- c("CA", "GT", "CG", "GC")
+  var_cols <- c(ts, tv1, tv2)
+  if (del) {
+    var_cols <- c(var_cols, dels)
+  }
+  return(var_cols)
+}
+
+#####################
+### Variant Plots ###
+#####################
+plot_variants <- function(full.t, out_dir, plot_prefix, fill_by) {
+  dir.create(path = out_dir, showWarnings = F)
+  # Per-Sample
+  tt <- full.t %>%
+    filter(grepl("chr", CHROM)) %>%
+    filter(!CHROM %in% c("chrY", "chrM")) %>%
+    group_by(Sample, !!!syms(fill_by)) %>%
+    summarise(Count = n())
+  write.xlsx(x = list("NumVariants" = tt), file = paste(out_dir, paste0(plot_prefix, "_VariantCounts.xlsx"), sep = "/"))
+  
+  p <- ggplot(tt, aes(x = Sample, y = Count, fill = get(fill_by), color = get(fill_by))) +
+    theme_bw(base_size = 12) +
+    theme(axis.text.x = element_text(angle = 90, hjust = 1, vjust = 0.5),
+          legend.position = "none") +
+    geom_bar(stat = "identity", alpha = 0.6) +
+    scale_y_continuous(labels = scales::comma, n.breaks = 8) +
+    xlab("") +
+    facet_grid(.~get(fill_by), scales = "free_x", space = "free")
+  ggsave(filename = paste(out_dir, paste0(plot_prefix, "_VariantCounts.pdf"), sep = "/"), plot = p, width = 6, height = 6)
+  
+  tt <- full.t %>%
+    filter(grepl("chr", CHROM)) %>%
+    filter(!CHROM %in% c("chrY", "chrM")) %>%
+    group_by(Sample, !!!syms(fill_by), REF, ALT) %>%
+    summarise(Count = n())
+  tt <- mutate(tt, TYPE = case_when(nchar(REF) == 1 & nchar(ALT) == 1 & ALT != "*" ~ "SNV",
+                                    nchar(REF) == 1 & nchar(ALT) == 1 & ALT == "*" ~ "DEL",
+                                    nchar(REF) > nchar(ALT) & nchar(REF) - nchar(ALT) >= 1 ~ "DEL",
+                                    nchar(REF) < nchar(ALT) & nchar(ALT) - nchar(REF) >= 1 ~ "INS",
+                                    TRUE ~ "Other"))
+  write.xlsx(x = list("VariantClass" = tt), file = paste(out_dir, paste0(plot_prefix, "_VariantClassification.xlsx"), sep = "/"))
+  tt_type <- tt %>%
+    group_by(Sample, !!!syms(fill_by), TYPE) %>%
+    summarise(SumCount = sum(Count)) %>%
+    arrange(desc(SumCount)) %>%
+    group_by(Sample, !!!syms(fill_by)) %>%
+    mutate(CountPerc =  SumCount/sum(SumCount))
+  write.xlsx(x = list("VariantType" = tt_type), file = paste(out_dir, paste0(plot_prefix, "_VariantType.xlsx"), sep = "/"))
+  
+  p <- ggplot(tt_type, aes(x = Sample, y = SumCount, fill = TYPE, color = TYPE)) +
+    theme_bw(base_size = 12) +
+    theme(axis.text.x = element_text(angle = 30, hjust = 1),
+          legend.position = "top") +
+    guides(fill = guide_legend(ncol = 6)) +
+    xlab("") +
+    ylab("Count") +
+    scale_fill_brewer(palette = "Set1", name = "") +
+    scale_color_brewer(palette = "Set1", name = "") +
+    geom_bar(stat = "identity", position = "stack", alpha = 0.6) +
+    scale_y_continuous(labels = scales::comma, n.breaks = 8) +
+    facet_grid(.~get(fill_by), scales = "free_x", space = "free")
+  ggsave(filename = paste(out_dir, paste0(plot_prefix, "_VariantTypes.pdf"), sep = "/"), plot = p, width = 6, height = 6)
+  
+  p <- ggplot(tt_type, aes(x = Sample, y = CountPerc, fill = TYPE, color = TYPE)) +
+    theme_bw(base_size = 12) +
+    theme(axis.text.x = element_text(angle = 30, hjust = 1),
+          legend.position = "top") +
+    guides(fill = guide_legend(ncol = 6)) +
+    xlab("") +
+    ylab("Count") +
+    scale_fill_brewer(palette = "Set1", name = "") +
+    scale_color_brewer(palette = "Set1", name = "") +
+    geom_bar(stat = "identity", position = "stack", alpha = 0.6) +
+    scale_y_continuous(labels = scales::percent_format(accuracy = 2), n.breaks = 10) +
+    facet_grid(.~get(fill_by), scales = "free_x", space = "free")
+  ggsave(filename = paste(out_dir, paste0(plot_prefix, "_VariantTypesPerc.pdf"), sep = "/"), plot = p, width = 6, height = 6)
+  
+  tt <- full.t %>%
+    filter(nchar(REF) == 1 & nchar(ALT) == 1 & grepl("chr", CHROM)) %>%
+    filter(!CHROM %in% c("chrY", "chrM")) %>%
+    group_by(Sample, !!!syms(fill_by), REF, ALT) %>%
+    summarise(Count = n()) %>%
+    group_by(Sample, !!!syms(fill_by)) %>%
+    mutate(CountPerc =  Count/sum(Count))
+  tt$Variant <- paste0(tt$REF, tt$ALT)
+  tt$Variant <- factor(tt$Variant, levels = sort(unique(tt$Variant)))
+  write.xlsx(x = list("SNV" = tt), file = paste(out_dir, paste0(plot_prefix, "_SNVcounts.xlsx"), sep = "/"))
+  
+  tt$Variant <- factor(tt$Variant, levels = rev(names(snv_colors2())))
+  p2 <- ggplot(tt, aes(x = Sample, y = Count, fill = Variant, color = Variant)) +
+    theme_bw(base_size = 12) +
+    theme(axis.text.x = element_text(angle = 30, hjust = 1)) +
+    geom_bar(stat = "identity", alpha = 0.6) +
+    scale_fill_manual(values = snv_colors2(del = F)) +
+    scale_color_manual(values = snv_colors2(del = F)) +
+    xlab("") +
+    scale_y_continuous(labels = scales::comma, n.breaks = 8) +
+    facet_grid(.~get(fill_by), scales = "free_x", space = "free")
+  ggsave(filename = paste(out_dir, paste0(plot_prefix, "_SNVcounts.pdf"), sep = "/"), plot = p2, width = 7, height = 7)
+  
+  p2p <- ggplot(tt, aes(x = Sample, y = CountPerc, fill = Variant, color = Variant)) +
+    theme_bw(base_size = 12) +
+    theme(axis.text.x = element_text(angle = 30, hjust = 1)) +
+    geom_bar(stat = "identity", alpha = 0.6) +
+    scale_fill_manual(values = snv_colors2(del = F)) +
+    scale_color_manual(values = snv_colors2(del = F)) +
+    scale_y_continuous(labels = scales::percent_format(accuracy = 2), n.breaks = 10) +
+    xlab("") +
+    facet_grid(.~get(fill_by), scales = "free_x", space = "free")
+  ggsave(filename = paste(out_dir, paste0(plot_prefix, "_SNVcountsPerc.pdf"), sep = "/"), plot = p2p, width = 7, height = 7)
+}
+
+
+###############
+### General ###
+###############
+# Change directory
+out_dir <- "."
+full.t.nomulti.noGerm <- read.xlsx(xlsxFile = paste(out_dir, "WES_results.xlsx", sep = "/"), sheet = "Full_NoMulti_noGerm")
+
+plot_variants(full.t = full.t.nomulti.noGerm, out_dir = out_dir, plot_prefix = "WES_results", fill_by = "Donor")
+
+snpeff_df <- full.t.nomulti.noGerm %>% filter(snpEff_Impact %in% c("HIGH", "MODERATE", "LOW")) %>%
+  group_by(Donor, Group, snpEff_Impact) %>% summarise(Count = n())
+snpeff_df <- melt(reshape2::dcast(snpeff_df, Donor + Group ~ snpEff_Impact, value.var = "Count"), id.vars = c("Donor", "Group"))
+snpeff_df$variable <- factor(snpeff_df$variable, levels = c("HIGH", "MODERATE", "LOW"))
+write.xlsx(x = list("VarClassification" = snpeff_df), file = paste(out_dir, "WES_results_Variant_Classification.xlsx", sep = "/"))
+
+p <- ggplot(snpeff_df, aes(x = Group, y = value, fill = variable, color = variable)) +
+  theme_bw(base_size = 14) +
+  theme(axis.text.x = element_text(angle = 45, hjust = 1)) +
+  geom_bar(stat = "identity", alpha = 0.6) +
+  ylab("Count") +
+  xlab("") +
+  ggtitle("Variant Classification") +
+  scale_fill_brewer(palette = "Set1", name = "Classification") +
+  scale_color_brewer(palette = "Set1", name = "Classification") +
+  facet_grid(.~Donor)
+ggsave(filename = paste(out_dir, "WES_results_Variant_Classification.pdf", sep = "/"),
+       plot = p, width = 8, height = 6)
+
+
+##################
+### Gene Lists ###
+##################
+full.circos <- read.xlsx(xlsxFile = paste(out_dir, "WES_results_GeneLists.xlsx", sep = "/"), sheet = "GeneLists_Variants")
+
+p <- ggplot(full.circos, aes(x = Sample, y = snpEff_GeneName, size = VAF, color = TYPE)) +
+  theme_bw(base_size = 10) +
+  theme(strip.text.y = element_text(angle = 0, hjust = 0, size = 12),
+        strip.background.y = element_rect(fill = "white", color = "white")) +
+  geom_jitter(height = 0, alpha = 0.5) +
+  scale_color_brewer(palette = "Set1", name = "Type") +
+  xlab("") +
+  ylab("") +
+  facet_grid(GL~Group, scales = "free", space = "free") +
+  guides(color = guide_legend(override.aes = list(size = 5)))
+ggsave(filename = paste(out_dir, paste0("WES_results_GeneLists.pdf"), sep = "/"),
+       plot = p, width = 10, height = 12)
+