Finish draft assembly track

bin/gen_report.R

@@ -0,0 +1,20 @@
+#!/usr/bin/env Rscript
+
+args <- commandArgs(trailingOnly = TRUE)
+workflow <- args[1]
+genome_size <- args[2]
+species_name <- args[3]
+
+# Generate rmarkdown report for the relevant track
+if (workflow == "draft") {
+    rmarkdown::render(
+      input  = 'report_draft_assembly.Rmd',
+      params = list(
+        quast_report = "transposed_report.tsv",
+        kraken_report = "kraken_reports.txt",
+        coverage_report = "coverage_reports.txt",
+        genome_size_val = genome_size,
+        species_name = species_name
+      )
+    )
+}

bin/generate_input.R

@@ -0,0 +1,89 @@
+#!/usr/bin/env Rscript
+
+args <- commandArgs(trailingOnly = TRUE)
+
+option <- args[1]
+
+if (option == "illumina") {
+    path <- args[2]
+    pattern <- args[3]
+
+    if (grepl("/$", path) == FALSE) {
+       path <- paste0(path, "/")
+    }
+
+    files <- list.files(path = path,
+                        pattern = pattern)
+
+    files_path <- paste0(path, files)
+
+    filenames <- unique(sub(pattern, "", files))
+    df <- data.frame(sample = filenames)
+    forward <- grep("_R1_", files_path, value = TRUE)
+    reverse <- grep("_R2_", files_path, value = TRUE)
+
+    df$R1 <- forward
+    df$R2 <- reverse
+
+    write.table(
+      x = df,
+      file = "samplesheet.csv",
+      quote = FALSE,
+      row.names = FALSE,
+      sep = ","
+    )
+
+} else if (option == "hybrid") {
+    path1 <- args[2]
+    path2 <- args[3]
+    pattern <- args[4]
+
+    if (grepl("/$", path1) == FALSE) {
+      path1 <- paste0(path1, "/")
+    }
+
+    if (grepl("/$", path2) == FALSE) {
+      path2 <- paste0(path2, "/")
+    }
+
+    files <- list.files(path = path1,
+                        pattern = pattern)
+
+    np_files <- list.files(path = path2,
+                           pattern = ".fastq.gz")
+
+    filenames <- sub(pattern, "", files)
+    filenames_np <- sub(".fastq.gz", "", np_files)
+
+    df <- data.frame(sample = filenames)
+    df$path <- paste0(path1, files)
+    df$read <- ifelse(grepl("_R1.fastq.gz", df$path), "R1", "R2")
+
+    df_np <- data.frame(sample = filenames_np,
+                        NP = paste0(path2, np_files))
+
+    il_samplesheet <- reshape(
+      data = df,
+      direction = "wide",
+      v.names = "path",
+      idvar = "sample",
+      timevar = "read"
+    )
+
+    samplesheet <- merge(
+      x = il_samplesheet,
+      y = df_np,
+      by = "sample",
+      all = TRUE
+    )
+
+    names(samplesheet) <- c("sample","R1","R2","NP")
+
+    write.table(
+      x         = samplesheet,
+      file      = "samplesheet.csv",
+      quote     = FALSE,
+      row.names = FALSE,
+      sep       = ","
+    )
+}

bin/merge_cov_data.R

@@ -0,0 +1,59 @@
+#!/usr/bin/env Rscript
+
+library(impoRt)
+library(dplyr)
+library(formattable)
+library(sparkline)
+
+# Import and merge coverage reports
+func_paste <- function(x, collapse = ",") paste(unique(x[!is.na(x)]), collapse = collapse)
+
+calc_cov <- function(df) {
+  df %>%
+    rename("reference" = X1,
+           "pos" = X2,
+           "reads" = X3) %>%
+    mutate(reads_no_unmapped = ifelse(reads == 0, NA, reads),
+           test = ifelse(reads == 0, 0, 1)) %>%
+    group_by(reference) %>%
+    summarise(reference = unique(reference),
+              cov_perc = round(sum(test)/max(pos)*100, 2),
+              mean_reads_per_base = round(mean(reads_no_unmapped, na.rm = TRUE), 2),
+              median_reads_per_base = median(reads_no_unmapped, na.rm = TRUE),
+              sd_reads_per_base = round(sd(reads_no_unmapped, na.rm = TRUE), 2)) %>%
+    ungroup() %>%
+    mutate(mean_cov = round(mean(mean_reads_per_base),2),
+           median_cov = median(median_reads_per_base),
+           mean_perc_cov = round(mean(cov_perc),2),
+           Coverage = spk_chr(mean_reads_per_base, width = 180, height = 25))
+}
+
+coverage_reports <- get_data(
+  ".",
+  pattern = "genomecov.txt",
+  col_names = FALSE,
+  df = FALSE
+) %>%
+  lapply(., calc_cov) %>%
+  bind_rows(.id = "sample") %>%
+  group_by(sample) %>%
+  summarise_all(list(func_paste)) %>%
+  mutate(sample = sub("_genomecov.txt", "", sample),
+         `Mean coverage` = color_bar(color = "#d9d9d9")(mean_cov),
+         `Median coverage` = color_bar(color = "#ccebc5")(median_cov),
+         `Mean percent coverage` = color_bar(color = "#fccde5")(mean_perc_cov)) %>%
+  select(sample,
+         `Mean coverage`,
+         `Median coverage`,
+         `Mean percent coverage`,
+         Coverage) %>%
+  rename("Assembly" = sample) %>%
+  mutate(Coverage = gsub("\n", "", Coverage))
+
+write.table(
+  coverage_reports,
+  "coverage_reports.txt",
+  sep = "\t",
+  quote = FALSE,
+  row.names = FALSE
+)

bin/merge_kraken_data.R

@@ -0,0 +1,28 @@
+#!/usr/bin/env Rscript
+
+library(impoRt)
+library(dplyr)
+
+# Import and merge kraken2 reports
+kraken_reports <- get_data(
+  ".",
+  pattern = "report",
+  col_names = FALSE,
+  convert = TRUE
+  ) %>%
+  rename("perc_fragments" = X1,
+         "num_fragments_root" = X2,
+         "num_fragments_direct" = X3,
+         "rank_code" = X4,
+         "NCBI_taxon" = X5,
+         "scientific_name" = X6) %>%
+  mutate_all(~trimws(., which = "both")) %>%
+  mutate(ref = sub("\\.kr2\\.report", "", ref))
+
+write.table(
+  kraken_reports,
+  "kraken_reports.txt",
+  sep = "\t",
+  quote = FALSE,
+  row.names = FALSE
+)