Upload tasks/amplicon-microbiome/scripts/05_report.R with huggingface_hub

tasks/amplicon-microbiome/scripts/05_report.R

@@ -0,0 +1,68 @@
+# Compile final report from all pipeline outputs
+
+# Read DADA2 summary
+dada2_summary <- readLines("outputs/dada2/summary.txt")
+parse_val <- function(lines, key) {
+  line <- grep(paste0("^", key, ":"), lines, value=TRUE)
+  trimws(sub(paste0("^", key, ": *"), "", line))
+}
+
+total_input <- parse_val(dada2_summary, "total_input_reads")
+total_filtered <- parse_val(dada2_summary, "total_filtered_reads")
+total_asvs <- parse_val(dada2_summary, "total_asvs")
+total_chimeras <- parse_val(dada2_summary, "total_chimeras")
+
+# Taxonomy
+tax <- read.table("outputs/taxonomy/taxonomy.tsv", header=TRUE, sep="\t", fill=TRUE)
+classified <- sum(!is.na(tax$Phylum))
+phylum_tab <- table(tax$Phylum, useNA="no")
+top_phylum <- names(sort(phylum_tab, decreasing=TRUE))[1]
+top_phylum_pct <- round(100 * max(phylum_tab) / sum(phylum_tab), 1)
+
+# Diversity
+div <- read.table("outputs/diversity/diversity_metrics.tsv", header=TRUE, sep="\t")
+mean_shannon <- div$value[div$metric == "mean_shannon"]
+mean_simpson <- div$value[div$metric == "mean_simpson"]
+mean_observed <- div$value[div$metric == "mean_observed_richness"]
+mean_pd <- div$value[div$metric == "mean_faith_pd"]
+
+# DA results
+da <- read.table("outputs/diversity/da_results.tsv", header=TRUE, sep="\t")
+da_features <- sum(da$padj < 0.05, na.rm=TRUE)
+
+# PICRUSt2 pathways
+pw_count <- 0
+pw_files <- c(
+  "outputs/picrust2/out/pathways_out/path_abun_unstrat.tsv.gz",
+  "outputs/picrust2/out/pathways_out/path_abun_unstrat.tsv"
+)
+for (pf in pw_files) {
+  if (file.exists(pf)) {
+    if (grepl("\\.gz$", pf)) {
+      pw <- read.table(gzfile(pf), header=TRUE, sep="\t", check.names=FALSE)
+    } else {
+      pw <- read.table(pf, header=TRUE, sep="\t", check.names=FALSE)
+    }
+    pw_count <- nrow(pw)
+    break
+  }
+}
+
+# Write report
+report <- data.frame(
+  metric = c("total_samples", "total_input_reads", "total_filtered_reads",
+             "total_sequence_variants", "chimeras_removed",
+             "classified_variants", "top_phylum", "top_phylum_pct",
+             "mean_shannon", "mean_simpson", "mean_observed_richness",
+             "mean_phylogenetic_diversity", "predicted_pathways",
+             "differentially_abundant_features"),
+  value = c(4, total_input, total_filtered, total_asvs, total_chimeras,
+            classified, top_phylum, top_phylum_pct,
+            mean_shannon, mean_simpson, mean_observed, mean_pd,
+            pw_count, da_features)
+)
+write.csv(report, "results/report.csv", row.names=FALSE, quote=FALSE)
+cat("=== Final Report ===\n")
+for (i in seq_len(nrow(report))) {
+  cat(sprintf("  %s = %s\n", report$metric[i], report$value[i]))
+}