Merge pull request #17 from umccr/seqrunsum_cli

Add FASTQ summary, fix colours, handle RNAsum v1

R/meta_rnasum.R

@@ -28,12 +28,30 @@ meta_rnasum <- function(pmeta, status = "Succeeded") {
     meta_io_fromjson() |>
     dplyr::mutate(
       # input
+      # renamed in v1.1.0
       gds_indir_dragen = purrr::map_chr(.data$input, list("dragen_transcriptome_directory", "location"), .default = NA),
+      gds_indir_dragen = ifelse(
+        is.na(gds_indir_dragen),
+        purrr::map_chr(.data$input, list("dragen_wts_dir", "location"), .default = NA),
+        gds_indir_dragen
+      ),
       gds_indir_umccrise = purrr::map_chr(.data$input, list("umccrise_directory", "location"), .default = NA),
+      # renamed in v1.1.0
       gds_indir_arriba = purrr::map_chr(.data$input, list("arriba_directory", "location"), .default = NA),
+      gds_indir_arriba = ifelse(
+        is.na(gds_indir_arriba),
+        purrr::map_chr(.data$input, list("arriba_dir", "location"), .default = NA),
+        gds_indir_arriba
+      ),
       rnasum_sample_name = purrr::map_chr(.data$input, "sample_name", .default = NA),
       rnasum_dataset = purrr::map_chr(.data$input, "dataset", .default = NA),
       rnasum_report_dir = purrr::map_chr(.data$input, "report_directory", .default = NA),
+      # renamed in v1.1.0
+      rnasum_report_dir = ifelse(
+        is.na(rnasum_report_dir),
+        purrr::map_chr(.data$input, "report_dir", .default = NA),
+        rnasum_report_dir
+      ),
       sbjid1 = sub("(SBJ.*)__L.*", "\\1", .data$rnasum_report_dir),
       libid1 = sub("(SBJ.*)__(L.*)", "\\2", .data$rnasum_report_dir),
       # output

R/meta_wgs_alignment_qc.R

@@ -10,7 +10,7 @@
 #'   readr::read_rds()
 #' (m <- meta_wgs_alignment_qc(pmeta))
 #' @testexamples
-#' expect_equal("Lane" %in% colnames(m), TRUE)
+#' expect_equal("lane" %in% colnames(m), TRUE)
 #' @export
 meta_wgs_alignment_qc <- function(pmeta, status = "Succeeded") {
   # retrieve workflow runs with the given type and status
@@ -27,22 +27,36 @@ meta_wgs_alignment_qc <- function(pmeta, status = "Succeeded") {
     meta_io_fromjson() |>
     dplyr::mutate(
       # input
-      rglb = purrr::map_chr(.data$input, list("fastq_list_rows", "rglb")),
+      rgid = purrr::map_chr(.data$input, list("fastq_list_rows", "rgid")),
       rgsm = purrr::map_chr(.data$input, list("fastq_list_rows", "rgsm")),
+      rglb = purrr::map_chr(.data$input, list("fastq_list_rows", "rglb")),
       lane = purrr::map_int(.data$input, list("fastq_list_rows", "lane")),
       lane = as.character(.data$lane),
+      # read_1/read_2 are dfs
+      fq1 = purrr::map_chr(.data$input, list("fastq_list_rows", "read_1", "location"), .default = NA),
+      fq2 = purrr::map_chr(.data$input, list("fastq_list_rows", "read_2", "location"), .default = NA),
       # output
       gds_outdir_dragen = purrr::map_chr(.data$output, list("dragen_alignment_output_directory", "location"), .default = NA),
       gds_outdir_multiqc = purrr::map_chr(.data$output, list("multiqc_output_directory", "location"), .default = NA),
       SubjectID = sub("umccr__automated__wgs_alignment_qc__(SBJ.*)__L.*", "\\1", .data$wfr_name),
+    ) |>
+    tidyr::separate_wider_delim(
+      cols = "rgid", delim = ".",
+      names = c("index1", "index2", "lane2", "illumina_id", "sample_lib_id")
     )
+
   d |>
     dplyr::select(
       dplyr::all_of(meta_main_cols()),
       "SubjectID",
       LibraryID = "rglb",
       SampleID = "rgsm",
-      Lane = "lane",
+      "lane",
+      "index1",
+      "index2",
+      "illumina_id",
+      "fq1",
+      "fq2",
       "gds_outdir_dragen",
       "gds_outdir_multiqc",
     )

R/meta_wts_alignment_qc.R

@@ -10,7 +10,7 @@
 #'   readr::read_rds()
 #' (m <- meta_wts_alignment_qc(pmeta))
 #' @testexamples
-#' expect_equal("Lane" %in% colnames(m), TRUE)
+#' expect_equal("lane" %in% colnames(m), TRUE)
 #' @export
 meta_wts_alignment_qc <- function(pmeta, status = "Succeeded") {
   # retrieve workflow runs with the given type and status
@@ -27,22 +27,36 @@ meta_wts_alignment_qc <- function(pmeta, status = "Succeeded") {
     meta_io_fromjson() |>
     dplyr::mutate(
       # input
-      rglb = purrr::map_chr(.data$input, list("fastq_list_rows", "rglb")),
+      rgid = purrr::map_chr(.data$input, list("fastq_list_rows", "rgid")),
       rgsm = purrr::map_chr(.data$input, list("fastq_list_rows", "rgsm")),
+      rglb = purrr::map_chr(.data$input, list("fastq_list_rows", "rglb")),
       lane = purrr::map_int(.data$input, list("fastq_list_rows", "lane")),
       lane = as.character(.data$lane),
+      # read_1/read_2 are dfs
+      fq1 = purrr::map_chr(.data$input, list("fastq_list_rows", "read_1", "location"), .default = NA),
+      fq2 = purrr::map_chr(.data$input, list("fastq_list_rows", "read_2", "location"), .default = NA),
       # output
       gds_outdir_dragen = purrr::map_chr(.data$output, list("dragen_alignment_output_directory", "location"), .default = NA),
       gds_outdir_multiqc = purrr::map_chr(.data$output, list("multiqc_output_directory", "location"), .default = NA),
       SubjectID = sub("umccr__automated__wts_alignment_qc__(SBJ.*)__L.*", "\\1", .data$wfr_name),
+    ) |>
+    tidyr::separate_wider_delim(
+      cols = "rgid", delim = ".",
+      names = c("index1", "index2", "lane2", "illumina_id", "sample_lib_id")
     )
+
   d |>
     dplyr::select(
       dplyr::all_of(meta_main_cols()),
       "SubjectID",
       LibraryID = "rglb",
       SampleID = "rgsm",
-      Lane = "lane",
+      "lane",
+      "index1",
+      "index2",
+      "illumina_id",
+      "fq1",
+      "fq2",
       "gds_outdir_dragen",
       "gds_outdir_multiqc",
     )

inst/reports/seqrunsum/funcs.R

@@ -1,4 +1,42 @@
 funcs <- list(
+  gds_fastqlistcsv_read = function(gdsdir, token, page_size = 20) {
+    .read_fastqlist <- function(x) {
+      nms <- tibble::tribble(
+        ~new_nm, ~old_nm, ~class,
+        "rgid", "RGID", "c",
+        "SampleID_LibraryID", "RGSM", "c",
+        "rglb", "RGLB", "c",
+        "lane", "Lane", "c",
+        "1", "Read1File", "c",
+        "2", "Read2File", "c"
+      )
+      lookup <- tibble::deframe(nms[c("new_nm", "old_nm")])
+      d <- readr::read_csv(x, col_types = readr::cols(.default = "c"))
+      assertthat::assert_that(all(colnames(d) == nms[["old_nm"]]))
+      d |>
+        dplyr::rename(dplyr::all_of(lookup)) |>
+        dplyr::mutate(
+          SampleID = sub("(.*)_(L.*)", "\\1", .data$SampleID_LibraryID),
+          LibraryID = sub("(.*)_(L.*)", "\\2", .data$SampleID_LibraryID),
+          topup = grepl("topup", .data$LibraryID)
+        ) |>
+        dplyr::select("rgid", "SampleID", "LibraryID", "lane", "1", "2", "topup") |>
+        tidyr::pivot_longer(c("1", "2"), names_to = "read", values_to = "path")
+    }
+    regex <- tibble::tribble(
+      ~regex, ~fun,
+      "fastq_list\\.csv$", "fastq_list"
+    )
+    g <- dracarys::gds_files_list_filter_relevant(
+      gdsdir = gdsdir, token = token, pattern = NULL, include_url = TRUE,
+      page_size = page_size, regexes = regex
+    )
+    assertthat::assert_that(
+      nrow(g) == 1,
+      all(colnames(g) == c("type", "bname", "size", "file_id", "path", "presigned_url"))
+    )
+    .read_fastqlist(g$presigned_url)
+  },
   #----#
   kable_empty_wf = function(wf) {
     kableExtra::kbl(NULL, caption = glue("<strong>NO {wf} WORKFLOWS WERE RUN</strong>"), escape = FALSE) |>
@@ -49,10 +87,10 @@ funcs <- list(
       ) |>
       ungroup() |>
       tidyr::unnest(clrs) |>
+      distinct(.data$clrs, .keep_all = TRUE) |>
       pull(clrs)
-    max_col <- length(clrs)
-    stopifnot(nc <= max_col, nc > 0)
-    clrs[seq_len(nc)]
+    # recycle colour vector according to nc
+    rep_len(clrs, length.out = nc)
   },
   #----#
   get_sbj_url = function(x, colour = NULL, account = "pro") {

inst/reports/seqrunsum/render.sh

@@ -1,9 +1,9 @@
-start="2024-05-18"
-end="2024-05-20"
-out="seqrunsum_${start}_${end}.html"
+date_start="2024-08-03"
+date_end="2024-08-05"
+out="seqrunsum_${date_start}_${date_end}.html"
 
 quarto render report.qmd \
-    -P date_start:${start} \
-    -P date_end:${end} \
+    -P date_start:${date_start} \
+    -P date_end:${date_end} \
     -o ${out} \
     --output-dir nogit