Merge pull request #26 from umccr/annual_summary

Annual workflow summary

.Rbuildignore

@@ -17,5 +17,8 @@
 ^nogit$
 ^vignettes$
 inst/reports/datashare/nogit
+inst/reports/mega_seqrunsum/.quarto
+inst/reports/mega_seqrunsum/nogit
+inst/reports/mega_seqrunsum/report_files
 inst/reports/seqrunsum/.quarto
 inst/reports/seqrunsum/nogit

DESCRIPTION

@@ -32,6 +32,7 @@ Imports:
     httr2,
     jose,
     jsonlite,
+    lubridate,
     optparse,
     paws,
     purrr,

R/meta_bcl_convert.R

@@ -61,9 +61,14 @@ meta_bcl_convert <- function(pmeta, status = "Succeeded") {
   d |>
     tidyr::separate_wider_regex("sample", c(sampleid = ".*", "_", libid1 = "L.*"), cols_remove = FALSE) |>
     tidyr::separate_wider_regex("libid1", c(libid2 = ".*", "_", topup_or_rerun = ".*"), cols_remove = FALSE, too_few = "align_start") |>
-    dplyr::mutate(gds_outdir_reports = file.path(dirname(.data$gds_outdir_multiqc), .data$batch_name, "Reports")) |>
+    dplyr::mutate(
+      gds_outdir_reports = file.path(dirname(.data$gds_outdir_multiqc), .data$batch_name, "Reports"),
+      year = as.character(lubridate::year(.data$start)),
+      durationMin = round(as.numeric(difftime(.data$end, .data$start, units = "mins")))
+    ) |>
     dplyr::select(
       dplyr::all_of(meta_main_cols()),
+      "year", "durationMin",
       -dplyr::any_of(c("batch_run")), # NA for bcl_convert
       SampleID = "sampleid",
       LibraryID = "libid2",

R/meta_oncoanalyser_wgs.R

@@ -36,11 +36,15 @@ meta_oncoanalyser_wgs <- function(pmeta, status = "Succeeded") {
       gds_bam_tumor = purrr::map_chr(.data$input, "tumor_wgs_bam", .default = NA),
       gds_bam_normal = purrr::map_chr(.data$input, "normal_wgs_bam", .default = NA),
       # output
-      s3_outdir_oncoanalyser = purrr::map_chr(.data$output, "output_directory", .default = NA)
+      s3_outdir_oncoanalyser = purrr::map_chr(.data$output, "output_directory", .default = NA),
+      # other
+      year = as.character(lubridate::year(.data$start)),
+      durationMin = round(as.numeric(difftime(.data$end, .data$start, units = "mins")))
     )
   d |>
     dplyr::select(
       dplyr::all_of(meta_main_cols()),
+      "year", "durationMin",
       "SubjectID",
       "LibraryID_tumor",
       "LibraryID_normal",

R/meta_oncoanalyser_wgts_existing_both.R

@@ -41,11 +41,15 @@ meta_oncoanalyser_wgts_existing_both <- function(pmeta, status = "Succeeded") {
       s3_indir_oncoanalyser_wgs = purrr::map_chr(.data$input, "existing_wgs_dir", .default = NA),
       s3_indir_oncoanalyser_wts = purrr::map_chr(.data$input, "existing_wts_dir", .default = NA),
       # output
-      s3_outdir_oncoanalyser = purrr::map_chr(.data$output, "output_directory", .default = NA)
+      s3_outdir_oncoanalyser = purrr::map_chr(.data$output, "output_directory", .default = NA),
+      # other
+      year = as.character(lubridate::year(.data$start)),
+      durationMin = round(as.numeric(difftime(.data$end, .data$start, units = "mins")))
     )
   d |>
     dplyr::select(
       dplyr::all_of(meta_main_cols()),
+      "year", "durationMin",
       "SubjectID",
       "LibraryID_tumor_wgs",
       "LibraryID_normal_wgs",

R/meta_oncoanalyser_wts.R

@@ -34,11 +34,15 @@ meta_oncoanalyser_wts <- function(pmeta, status = "Succeeded") {
       LibraryID = purrr::map_chr(.data$input, "tumor_wts_library_id", .default = NA),
       s3_bam = purrr::map_chr(.data$input, "tumor_wts_bam", .default = NA),
       # output
-      s3_outdir_oncoanalyser = purrr::map_chr(.data$output, "output_directory", .default = NA)
+      s3_outdir_oncoanalyser = purrr::map_chr(.data$output, "output_directory", .default = NA),
+      # other
+      year = as.character(lubridate::year(.data$start)),
+      durationMin = round(as.numeric(difftime(.data$end, .data$start, units = "mins")))
     )
   d |>
     dplyr::select(
       dplyr::all_of(meta_main_cols()),
+      "year", "durationMin",
       "SubjectID",
       "LibraryID",
       "SampleID",

R/meta_rnasum.R

@@ -63,11 +63,15 @@ meta_rnasum <- function(pmeta, status = "Succeeded") {
       # output
       gds_outfile_rnasum_html = purrr::map_chr(.data$output, list("rnasum_html", "location"), .default = NA),
       gds_outdir_rnasum = purrr::map_chr(.data$output, list("rnasum_output_directory", "location"), .default = NA),
+      # other
+      year = as.character(lubridate::year(.data$start)),
+      durationMin = round(as.numeric(difftime(.data$end, .data$start, units = "mins")))
     )
   d |>
     dplyr::select(
       dplyr::all_of(meta_main_cols()),
       -dplyr::any_of(c("sequence_run", "batch_run")), # NA for rnasum
+      "year", "durationMin",
       SubjectID = "sbjid1",
       LibraryID = "libid1",
       SampleID = "rnasum_sample_name",

R/meta_sash.R

@@ -37,11 +37,15 @@ meta_sash <- function(pmeta, status = "Succeeded") {
       gds_indir_dragen_germline = purrr::map_chr(.data$input, "dragen_germline_dir", .default = NA),
       s3_indir_oncoanalyser = purrr::map_chr(.data$input, "oncoanalyser_dir", .default = NA),
       # output
-      s3_outdir_sash = purrr::map_chr(.data$output, "output_directory", .default = NA)
+      s3_outdir_sash = purrr::map_chr(.data$output, "output_directory", .default = NA),
+      # other
+      year = as.character(lubridate::year(.data$start)),
+      durationMin = round(as.numeric(difftime(.data$end, .data$start, units = "mins")))
     )
   d |>
     dplyr::select(
       dplyr::all_of(meta_main_cols()),
+      "year", "durationMin",
       "SubjectID",
       "LibraryID_tumor",
       "LibraryID_normal",

R/meta_star_alignment.R

@@ -34,11 +34,15 @@ meta_star_alignment <- function(pmeta, status = "Succeeded") {
       gds_fq_fwd = purrr::map_chr(.data$input, "fastq_fwd", .default = NA),
       gds_fq_rev = purrr::map_chr(.data$input, "fastq_rev", .default = NA),
       # output
-      s3_outdir_star = purrr::map_chr(.data$output, "output_directory", .default = NA)
+      s3_outdir_star = purrr::map_chr(.data$output, "output_directory", .default = NA),
+      # other
+      year = as.character(lubridate::year(.data$start)),
+      durationMin = round(as.numeric(difftime(.data$end, .data$start, units = "mins")))
     )
   d |>
     dplyr::select(
       dplyr::all_of(meta_main_cols()),
+      "year", "durationMin",
       "SubjectID",
       "LibraryID",
       "SampleID",

R/meta_tso_ctdna_tumor_only.R

@@ -35,11 +35,15 @@ meta_tso_ctdna_tumor_only <- function(pmeta, status = c("Succeeded")) {
       libid1 = sub(".*_(L.*)", "\\1", .data$sample_id),
       rerun = grepl("rerun", .data$libid1),
       subjectid = sub("umccr__automated__tso_ctdna_tumor_only__(SBJ.*)__L.*", "\\1", .data$wfr_name),
-      libid = sub("umccr__automated__tso_ctdna_tumor_only__SBJ.*__(L.*)__.*", "\\1", .data$wfr_name) # equal to libid1 wo _rerun
+      libid = sub("umccr__automated__tso_ctdna_tumor_only__SBJ.*__(L.*)__.*", "\\1", .data$wfr_name), # equal to libid1 wo _rerun
+      # other
+      year = as.character(lubridate::year(.data$start)),
+      durationMin = round(as.numeric(difftime(.data$end, .data$start, units = "mins")))
     )
   d |>
     dplyr::select(
       dplyr::all_of(meta_main_cols()),
+      "year", "durationMin",
       SubjectID = "subjectid",
       LibraryID = "libid",
       SampleID = "sample_name2",

R/meta_umccrise.R

@@ -76,12 +76,16 @@ meta_umccrise <- function(pmeta, status = "Succeeded") {
       gds_outdir_umccrise1 = purrr::map_chr(.data$output, list("umccrise_output_directory", "location"), .default = NA),
       gds_outdir_umccrise = dplyr::if_else(
         is.na(.data$gds_outdir_umccrise1), .data$gds_outdir_umccrise2, .data$gds_outdir_umccrise1
-      )
+      ),
+      # other
+      year = as.character(lubridate::year(.data$start)),
+      durationMin = round(as.numeric(difftime(.data$end, .data$start, units = "mins")))
     )
   d |>
     dplyr::select(
       dplyr::all_of(meta_main_cols()),
       -dplyr::any_of(c("sequence_run", "batch_run")), # NA for umccrise
+      "year", "durationMin",
       "SubjectID",
       "LibraryID_tumor",
       "LibraryID_normal",

R/meta_wgs_alignment_qc.R

@@ -39,6 +39,9 @@ meta_wgs_alignment_qc <- function(pmeta, status = "Succeeded") {
       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__.*__wgs_alignment_qc__(SBJ.*)__L.*", "\\1", .data$wfr_name),
+      # other
+      year = as.character(lubridate::year(.data$start)),
+      durationMin = round(as.numeric(difftime(.data$end, .data$start, units = "mins")))
     ) |>
     tidyr::separate_wider_delim(
       cols = "rgid", delim = ".",
@@ -48,6 +51,7 @@ meta_wgs_alignment_qc <- function(pmeta, status = "Succeeded") {
   d |>
     dplyr::select(
       dplyr::all_of(meta_main_cols()),
+      "year", "durationMin",
       "SubjectID",
       LibraryID = "rglb",
       SampleID = "rgsm",

R/meta_wgs_tumor_normal.R

@@ -60,12 +60,21 @@ meta_wgs_tumor_normal <- function(pmeta, status = "Succeeded") {
       gds_outfile_dragen_somatic_snv_vcf = purrr::map_chr(.data$output, list("somatic_snv_vcf_out", "location"), .default = NA),
       gds_outfile_dragen_somatic_snv_vcf_hardfilt = purrr::map_chr(.data$output, list("somatic_snv_vcf_hard_filtered_out", "location"), .default = NA),
       gds_outfile_dragen_somatic_sv_vcf = purrr::map_chr(.data$output, list("somatic_structural_vcf_out", "location"), .default = NA),
-      SubjectID = sub("umccr__automated__wgs_tumor_normal__(SBJ.....)__L.*", "\\1", .data$wfr_name) # infer from wfr name
+      SubjectID = sub("umccr__automated__wgs_tumor_normal__(SBJ.....)__L.*", "\\1", .data$wfr_name),
+      SubjectID = ifelse(
+        !grepl("external_apgi", .data$wfr_name),
+        .data$SubjectID,
+        sub("umccr__external_apgi__wgs_tumor_normal__(.*)", "\\1", .data$wfr_name)
+      ),
+      # other
+      year = as.character(lubridate::year(.data$start)),
+      durationMin = round(as.numeric(difftime(.data$end, .data$start, units = "mins")))
     )
   d |>
     dplyr::select(
       dplyr::all_of(meta_main_cols()),
       -dplyr::any_of(c("sequence_run", "batch_run")), # NA for wgs_tumor_normal
+      "year", "durationMin",
       "SubjectID",
       "LibraryID_tumor",
       "LibraryID_normal",

R/meta_wts_alignment_qc.R

@@ -39,6 +39,9 @@ meta_wts_alignment_qc <- function(pmeta, status = "Succeeded") {
       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__.*__wts_alignment_qc__(SBJ.*)__L.*", "\\1", .data$wfr_name),
+      # other
+      year = as.character(lubridate::year(.data$start)),
+      durationMin = round(as.numeric(difftime(.data$end, .data$start, units = "mins")))
     ) |>
     tidyr::separate_wider_delim(
       cols = "rgid", delim = ".",
@@ -48,6 +51,7 @@ meta_wts_alignment_qc <- function(pmeta, status = "Succeeded") {
   d |>
     dplyr::select(
       dplyr::all_of(meta_main_cols()),
+      "year", "durationMin",
       "SubjectID",
       LibraryID = "rglb",
       SampleID = "rgsm",

R/meta_wts_tumor_only.R

@@ -28,20 +28,29 @@ meta_wts_tumor_only <- function(pmeta, status = "Succeeded") {
     meta_io_fromjson() |>
     dplyr::mutate(
       # input
-      rglb = purrr::map_chr(.data$input, \(x) unique(x[["fastq_list_rows"]][["rglb"]])),
-      rgsm = purrr::map_chr(.data$input, \(x) unique(x[["fastq_list_rows"]][["rgsm"]])),
+      rglb = purrr::map_chr(.data$input, \(x) unique(x[["fastq_list_rows"]][["rglb"]]) %||% NA),
+      rgsm = purrr::map_chr(.data$input, \(x) unique(x[["fastq_list_rows"]][["rgsm"]]) %||% NA),
       lane = purrr::map_chr(.data$input, \(x) paste(x[["fastq_list_rows"]][["lane"]], collapse = ",")),
       lane = as.character(.data$lane),
       # output
       gds_outdir_dragen = purrr::map_chr(.data$output, list("dragen_transcriptome_output_directory", "location"), .default = NA),
       gds_outdir_multiqc = purrr::map_chr(.data$output, list("multiqc_output_directory", "location"), .default = NA),
       gds_outdir_arriba = purrr::map_chr(.data$output, list("arriba_output_directory", "location"), .default = NA),
       gds_outdir_qualimap = purrr::map_chr(.data$output, list("qualimap_output_directory", "location"), .default = NA),
-      SubjectID = sub("umccr__.*__wts_tumor_only__(SBJ.*)__L.*", "\\1", .data$wfr_name)
+      SubjectID = sub("umccr__.*__wts_tumor_only__(SBJ.*)__L.*", "\\1", .data$wfr_name),
+      SubjectID = ifelse(
+        !grepl("external_apgi", .data$wfr_name),
+        .data$SubjectID,
+        sub("umccr__external_apgi__wts_tumor_only__(.*)", "\\1", .data$wfr_name)
+      ),
+      # other
+      year = as.character(lubridate::year(.data$start)),
+      durationMin = round(as.numeric(difftime(.data$end, .data$start, units = "mins")))
     )
   d |>
     dplyr::select(
       dplyr::all_of(meta_main_cols()),
+      "year", "durationMin",
       "SubjectID",
       LibraryID = "rglb",
       SampleID = "rgsm",

conda/recipe/meta.yaml

@@ -29,6 +29,7 @@ requirements:
     - r-httr2
     - r-jose
     - r-jsonlite
+    - r-lubridate
     - r-optparse
     - r-paws
     - r-purrr
@@ -52,6 +53,7 @@ requirements:
     - r-httr2
     - r-jose
     - r-jsonlite
+    - r-lubridate
     - r-optparse
     - r-paws
     - r-purrr

inst/reports/mega_seqrunsum/.gitignore

@@ -0,0 +1,4 @@
+/.quarto/
+nogit
+*html
+report_files

inst/reports/mega_seqrunsum/README.md

@@ -0,0 +1,13 @@
+# Mega Sequencing Run Summary
+
+Same as a typical Sequencing Run Summary, but for a large number of samples over a long period of time.
+
+**Contents**
+
+- Visualisation of workflow runtimes based on the PortalDB Workflow table.
+- Summary of sample metadata and workflow input/output paths.
+
+**Inputs**
+
+- PortalDB `workflow` slice for a given timeframe.
+- PortalDB `limsrow` slice for a given set of library IDs.

inst/reports/mega_seqrunsum/_quarto.yml

@@ -0,0 +1,2 @@
+project:
+  title: "Sequencing Run Summary"

inst/reports/mega_seqrunsum/funcs.R

@@ -0,0 +1,43 @@
+funcs <- list(
+  #----#
+  dt_view = function(x, id, height = 500, ...) {
+    htmltools::browsable(
+      htmltools::tagList(
+        htmltools::tags$button(
+          htmltools::tagList(fontawesome::fa("download"), "CSV"),
+          onclick = glue("Reactable.downloadDataCSV('{id}', '{id}.csv')")
+        ),
+        x |>
+          reactable::reactable(
+            bordered = TRUE,
+            filterable = TRUE,
+            fullWidth = TRUE,
+            height = height,
+            highlight = TRUE,
+            pagination = FALSE,
+            resizable = TRUE,
+            searchable = TRUE,
+            sortable = TRUE,
+            striped = TRUE,
+            wrap = FALSE,
+            elementId = id,
+            ...
+          )
+      )
+    )
+  },
+  func_eval = function(f) {
+    eval(parse(text = f))
+  },
+  #----#
+  get_ids = function(d, id) {
+    .get_ids <- function(tbl, id) {
+      tbl |>
+        select(contains(id)) |>
+        unlist() |>
+        unique()
+    }
+    d |>
+      mutate(ids = list(.get_ids(.data$tidy_meta, {{ id }})))
+  }
+)

inst/reports/mega_seqrunsum/render.sh

@@ -0,0 +1,5 @@
+out="mega_seqrunsum.html"
+
+quarto render report.qmd \
+    -o ${out} \
+    --output-dir nogit