WIP: Case study report automation

.gitignore

@@ -1,4 +1,5 @@
 .Rproj.user
 docs
+common_data
 .Rhistory
-.DS_Store
+.DS_Store

R/ipr2viz.R

@@ -94,14 +94,14 @@ getTopAccByLinDomArch <- function(infile_full,
     cln_domarch <- cln %>% select(domarch_cols)
     col_counts <- colSums(is.na(cln_domarch))
     DA_sym <- sym(names(which.min(col_counts)))
-    showNotification(paste0("Selecting representatives by unique ", DA_sym, " and lineage combinations"))
+    # showNotification(paste0("Selecting representatives by unique ", DA_sym, " and lineage combinations"))
     ## Group by Lineage, DomArch and reverse sort by group counts
     grouped <- cln %>%
         group_by({{ DA_sym }}, {{ lin_sym }}) %>%
         arrange(desc(PcPositive)) %>%
         summarise(count = n(), AccNum = dplyr::first(AccNum)) %>%
         arrange(-count) %>%
-        filter({{ lin_sym }} != "" && {{ DA_sym }} != "")
+        filter({{ lin_sym }} != "" & {{ DA_sym }} != "")
     top_acc <- grouped$AccNum[1:n]
     top_acc <- na.omit(top_acc)
     return(top_acc)
@@ -180,7 +180,7 @@ plotIPR2Viz <- function(infile_ipr = NULL, infile_full = NULL, accessions = c(),
     ipr_out <- read_tsv(infile_ipr, col_names = T, col_types = MolEvolvR::iprscan_cols)
     ipr_out <- ipr_out %>% filter(.data$Name %in% accessions)
     analysis_cols <- paste0("DomArch.", analysis)
-    infile_full <- infile_full %>% select(.data$analysis_cols, .data$Lineage_short, .data$QueryName, .data$PcPositive, .data$AccNum)
+    infile_full <- infile_full %>% select(analysis_cols, .data$Lineage_short, .data$QueryName, .data$PcPositive, .data$AccNum)
     ## To filter by Analysis
     analysis <- paste(analysis, collapse = "|")
     ## @SAM: This can't be set in stone since the analysis may change!
@@ -212,7 +212,7 @@ plotIPR2Viz <- function(infile_ipr = NULL, infile_full = NULL, accessions = c(),
     analysis_labeler <- analyses %>%
         pivot_wider(names_from = .data$Analysis, values_from = .data$Analysis)
 
-    lookup_tbl_path <- "/data/research/jravilab/common_data/cln_lookup_tbl.tsv"
+    system.file("common_data", "cln_lookup_tbl.tsv", package = "MolEvolvR", mustWork = TRUE)
     lookup_tbl <- read_tsv(lookup_tbl_path, col_names = T, col_types = MolEvolvR::lookup_table_cols)
 
     lookup_tbl <- lookup_tbl %>% select(-.data$ShortName) # Already has ShortName -- Just needs SignDesc

R/networks_domarch.R

@@ -211,7 +211,7 @@ createDomainNetwork <- function(prot, column = "DomArch", domains_of_interest, c
                 visOptions(highlightNearest = TRUE)
         },
         error = function(e) {
-            showNotification(toString(e))
+            # showNotification(toString(e))
             vis_g <- "error"
         },
         finally = {
@@ -364,4 +364,4 @@ createBinaryDomainNetwork <- function(prot, column = "DomArch", domains_of_inter
         "auto" = visIgraphLayout(vg, "layout.auto")
     )
     vg
-}
+}

R/pre-msa-tree.R

@@ -676,7 +676,7 @@ acc2FA <- function(accessions, outpath, plan = "sequential") {
                 id = accessions_partitioned[[x]],
                 db = "protein",
                 rettype = "fasta",
-                api_key = Sys.getenv("ENTREZ_API_KEY")
+                # api_key = Sys.getenv("ENTREZ_API_KEY")
             )
         )
     })
@@ -809,9 +809,9 @@ alignFasta <- function(fasta_file, tool = "Muscle", outpath = NULL) {
         abort("Error: The FASTA file does not exist: ", fasta_file)
     }
 
-    if (file_ext(fasta_file) != "fasta" && file_ext(fasta_file) != "fa") {
-        abort("Error: The specified file is not a valid FASTA file: ", fasta_file)
-    }
+    # if (file_ext(fasta_file) != "fasta" && file_ext(fasta_file) != "fa") {
+    #    abort("Error: The specified file is not a valid FASTA file: ", fasta_file)
+    # }
     fasta <- readAAStringSet(fasta_file)
 
     aligned <- switch(tool,
@@ -869,11 +869,11 @@ writeMSA_AA2FA <- function(alignment, outpath) {
         abort("Error: The output directory does not exist: ", outdir)
     }
 
-    l <- length(rownames(alignment))
+    l <- length(names(unmasked((alignment))))
     fasta <- ""
     for (i in 1:l)
     {
-        fasta <- paste0(fasta, paste(">", rownames(alignment)[i]), "\n")
+        fasta <- paste0(fasta, paste(">", names(unmasked((alignment)))[i]), "\n")
         seq <- toString(unmasked(alignment)[[i]])
         fasta <- paste0(fasta, seq, "\n")
     }
@@ -955,4 +955,4 @@ getAccNumFromFA <- function(fasta_file) {
 #   cfile <- read_delim("data/alignments/pspc.gismo.aln", delim=" ")
 #   cfile <- as.data.frame(map(cfile,function(x) gsub("\\s+", "",x)))
 #   colnames(cfile) <- c("AccNum", "Alignment")
-# }
+# }

R/summarize.R

@@ -690,18 +690,18 @@ totalGenContextOrDomArchCounts <- function(prot, column = "DomArch", lineage_col
         abort("Error: 'digits' must be a non-negative integer.")
     }
 
-    column <- sym(column)
+    # column <- sym(column)
 
     prot <- select(prot, {{ column }}, {{ lineage_col }}) %>%
         filter(!is.na({{ column }}) & !is.na({{ lineage_col }})) %>%
         filter({{ column }} != "")
 
     prot <- summarizeByLineage(prot, column, by = lineage_col, query = "all")
     col_count <- prot %>%
-        group_by({{ column }}) %>%
+        group_by(!!sym(column)) %>%
         summarise(totalcount = sum(count))
 
-    total <- left_join(prot, col_count, by = as_string(column))
+    total <- left_join(prot, col_count, by = column)
 
     sum_count <- sum(total$count)
     total <- total %>%
@@ -901,4 +901,4 @@ findParalogs <- function(prot) {
 # cat("Word counts for broken up domains from DAs and DAs from GCs.
 # \nFor e.g.:
 # DA.doms.wc <- query.sub$DA.doms %>%
-#   words2WordCounts()")
+#   words2WordCounts()")