Merge pull request #39 from alephnull7/master

Version 1.0.11
michaelgruenstaeudl · Apr 8, 2024 · de6eebf · de6eebf
2 parents b102b44 + e76876b
commit de6eebf
Show file tree

Hide file tree

Showing 17 changed files with 174 additions and 129 deletions.
diff --git a/CHANGELOG.md b/CHANGELOG.md
@@ -1,12 +1,22 @@
 CHANGELOG
 ---------
 
-#### Version 1.0.10 (2024.03.22)
+#### Version 1.0.11 (2024.03.27)
+* Minor fixes to tabular statistics file creation
+* A coverage of 0 for a window is always labeled as low coverage
+
+#### Version 1.0.10 (2024.03.23)
 * Dynamic changes to `IRCheck` as `PACVr.complete()` progresses
   * If no samples from `gbkFile` contain the qualifiers needed for specified analysis, continues with `IRCheck = NA`
   * If no IR regions are identified, continues with `IRCheck = NA`
   * If there are missing IR regions needed for synteny test, continues with `IRCheck = 0`
   * If no synteny is found, continues with `IRCheck = 0`
+  * When `tabularCovStats = TRUE`, if IR mismatches are identified, continues with `IRCheck = NA`
+* For `tabularCovStats = TRUE`, summary data corresponding to the complete genome is included in `<sampleName>_summary.regions`
+  * `<sampleName>_coverage.summary.regions` has been renamed `<sampleName>_summary.regions`
+  * The number of ambiguous nucleotides is always counted, named `N_count`
+  * When synteny testing is indicated by `IRCheck` and possible, as detailed above, the number of mismatches between the IRs is counted as `IR_mismatches`
+* The evenness metric for `tabularCovStats` has been renamed `E_score`
 
 #### Version 1.0.9 (2024.03.01)
 * Improving variable and file names

diff --git a/DESCRIPTION b/DESCRIPTION
@@ -1,6 +1,6 @@
 Package: PACVr
-Version: 1.0.10
-Date: 2024-03-22
+Version: 1.0.11
+Date: 2024-03-27
 Title: Plastome Assembly Coverage Visualization
 Authors@R: c(person("Gregory", "Smith", role=c("ctb")),
              person("Nils", "Jenke", role=c("ctb")),

diff --git a/R/AnalysisSpecs.R b/R/AnalysisSpecs.R
@@ -1,7 +1,7 @@
 #!/usr/bin/env RScript
 #contributors=c("Gregory Smith", "Nils Jenke", "Michael Gruenstaeudl")
 #email="m_gruenstaeudl@fhsu.edu"
-#version="2024.03.22.1637"
+#version="2024.03.27.0311"
 
 AnalysisSpecs <- R6Class("AnalysisSpecs",
   public = list(
@@ -67,6 +67,10 @@ AnalysisSpecs <- R6Class("AnalysisSpecs",
         self$regions_name <- "Chromosome"
         self$regions_start <- "chromStart"
         self$regions_end <- "chromEnd"
+      } else {
+        self$regions_name <- "Source"
+        self$regions_start <- "srcStart"
+        self$regions_end <- "srcEnd"
       }
     }
   )

diff --git a/R/GBKData.R b/R/GBKData.R
@@ -1,7 +1,7 @@
 #!/usr/bin/env RScript
 #contributors=c("Gregory Smith", "Nils Jenke", "Michael Gruenstaeudl")
 #email="m_gruenstaeudl@fhsu.edu"
-#version="2024.03.22.1637"
+#version="2024.03.27.0311"
 
 GBKData <- R6Class("GBKData",
   public = list(

diff --git a/R/IROps.R b/R/IROps.R
@@ -1,7 +1,7 @@
 #!/usr/bin/env RScript
 #contributors=c("Gregory Smith", "Nils Jenke", "Michael Gruenstaeudl")
 #email="m_gruenstaeudl@fhsu.edu"
-#version="2024.03.22.1637"
+#version="2024.03.27.0311"
 
 checkIREquality <- function(gbkData,
                             analysisSpecs) {
@@ -24,7 +24,7 @@ checkIREquality <- function(gbkData,
     IR_mismatches <- 0
   }
 
-  if (IR_mismatches > 0) {
+  if (IR_mismatches < 0) {
     logger::log_warn(
       "Proceeding with coverage depth visualization, but without quadripartite genome structure ..."
     )

diff --git a/R/OutputSpecs.R b/R/OutputSpecs.R
@@ -1,7 +1,7 @@
 #!/usr/bin/env RScript
 #contributors=c("Gregory Smith", "Nils Jenke", "Michael Gruenstaeudl")
 #email="m_gruenstaeudl@fhsu.edu"
-#version="2024.03.22.1637"
+#version="2024.03.27.0311"
 
 OutputSpecs <- R6Class("OutputSpecs",
   public = list(

diff --git a/R/PACVr.R b/R/PACVr.R
@@ -1,7 +1,7 @@
 #!/usr/bin/env RScript
 #contributors=c("Gregory Smith", "Nils Jenke", "Michael Gruenstaeudl")
 #email="m_gruenstaeudl@fhsu.edu"
-#version="2024.03.22.1637"
+#version="2024.03.27.0311"
 
 PACVr.read.gb <- function(gbkFile) {
   gbkRaw <- getGbkRaw(gbkFile)

diff --git a/R/RCircosOps.R b/R/RCircosOps.R
@@ -1,7 +1,7 @@
 #!/usr/bin/env RScript
 #contributors=c("Gregory Smith", "Nils Jenke", "Michael Gruenstaeudl")
 #email="m_gruenstaeudl@fhsu.edu"
-#version="2024.03.22.1637"
+#version="2024.03.27.0311"
 
 
 # The following R functions were taken from the R package RCircos and then modified.

diff --git a/R/compileStats.R b/R/compileStats.R
@@ -1,7 +1,7 @@
 #!/usr/bin/env RScript
 #contributors=c("Gregory Smith", "Nils Jenke", "Michael Gruenstaeudl")
 #email="m_gruenstaeudl@fhsu.edu"
-#version="2024.03.22.1637"
+#version="2024.03.27.0311"
 
 PACVr.compileCovStats <- function(gbkData,
                                   coverageRaw,

diff --git a/R/coverageCalcOps.R b/R/coverageCalcOps.R
@@ -1,7 +1,7 @@
 #!/usr/bin/env RScript
 #contributors=c("Gregory Smith", "Nils Jenke", "Michael Gruenstaeudl")
 #email="m_gruenstaeudl@fhsu.edu"
-#version="2024.03.22.1637"
+#version="2024.03.27.0311"
 
 CovCalc <- function(coverageRaw,
                     windowSize = 250,
@@ -126,39 +126,34 @@ filter_IR_regions <- function(coverageRaw, seqnames, covData, analysisSpecs) {
   return(covData)
 }
 
-setLowCoverage <- function(covData, analysisSpecs) {
-  # ir_regions
-  ir_regions <- covData$ir_regions
-  regions_name <- analysisSpecs$regions_name
-  aggFormula <- stats::as.formula(paste("coverage ~", regions_name))
-  cov_regions <-
-    aggregate(
-      aggFormula,
-      data = ir_regions,
-      FUN = function(x)
-        ceiling(mean(x) - sd(x))
-    )
-  ir_regions$lowCoverage <- ir_regions$coverage < cov_regions$coverage[match(ir_regions[[regions_name]],
-                                                                             cov_regions[[regions_name]])]
-  ir_regions$lowCoverage[ir_regions$lowCoverage == TRUE] <- "*"
-  ir_regions$lowCoverage[ir_regions$lowCoverage == FALSE] <- ""
-  covData$ir_regions <- ir_regions
-
-  # ir_genes
-  ir_genes <- covData$ir_genes
-  ir_genes$lowCoverage <- ir_genes$coverage < mean(ir_genes$coverage) - sd(ir_genes$coverage)
-  ir_genes$lowCoverage[ir_genes$lowCoverage == TRUE] <- "*"
-  ir_genes$lowCoverage[ir_genes$lowCoverage == FALSE] <- ""
-  covData$ir_genes <- ir_genes
+setLowCoverages <- function(covData, analysisSpecs) {
+  covData$ir_regions <- setLowCoverage(covData$ir_regions,
+                                       analysisSpecs$regions_name)
+  covData$ir_genes <- setLowCoverage(covData$ir_genes)
+  covData$ir_noncoding <- setLowCoverage(covData$ir_noncoding)
+  return(covData)
+}
 
-  # ir_noncoding
-  ir_noncoding <- covData$ir_noncoding
-  ir_noncoding$lowCoverage <- ir_noncoding$coverage < mean(ir_noncoding$coverage) - sd(ir_noncoding$coverage)
-  ir_noncoding$lowCoverage[ir_noncoding$lowCoverage == TRUE] <- "*"
-  ir_noncoding$lowCoverage[ir_noncoding$lowCoverage == FALSE] <- ""
-  covData$ir_noncoding <- ir_noncoding
+setLowCoverage <- function(covDataField, regions_name = NULL) {
+  if (!is.null(regions_name)) {
+    aggFormula <- stats::as.formula(paste("coverage ~", regions_name))
+    cov_regions <-
+      aggregate(
+        aggFormula,
+        data = covDataField,
+        FUN = function(x)
+          ceiling(mean(x) - sd(x))
+      )
+    lowThreshold <- cov_regions$coverage[match(covDataField[[regions_name]],
+                                               cov_regions[[regions_name]])]
+  } else {
+    lowThreshold <- mean(covDataField$coverage) - sd(covDataField$coverage)
+  }
 
-  return(covData)
+  covDataField$lowCoverage <- (covDataField$coverage < lowThreshold) |
+                                (covDataField$coverage == 0)
+  covDataField$lowCoverage <- ifelse(covDataField$lowCoverage, "*", "")
+  return(covDataField)
 }
 
 # adapted from `nilsj9/PlastidSequenceCoverage`
@@ -182,6 +177,10 @@ updateRegionsSummary <- function(covSummaries,
   covSumRegions <- covSummaries$regions_summary
   regions_evenness <- getCovEvenness(covDataRegions,
                                      regions_name)
+  if (regions_name == "Source") {
+    covSumRegions[regions_name] <- "Complete_genome"
+    regions_evenness[regions_name] <- "Complete_genome"
+  }
   covSumRegions <- dplyr::full_join(covSumRegions,
                                     regions_evenness,
                                     regions_name)
@@ -257,6 +256,9 @@ getCovDepth <- function(covDataField, regions_name = NULL) {
   }
   covDepth <- covDataField %>%
     calcCovDepth()
+  if (!is.null(regions_name) && regions_name == "Source") {
+    covDepth[regions_name] <- "Unpartitioned"
+  }
   return(covDepth)
 }
 

diff --git a/R/customRead.gb.R b/R/customRead.gb.R
@@ -1,7 +1,7 @@
 #!/usr/bin/env RScript
 #contributors=c("Gregory Smith", "Nils Jenke", "Michael Gruenstaeudl")
 #email="m_gruenstaeudl@fhsu.edu"
-#version="2024.03.22.1637"
+#version="2024.03.27.0311"
 
 read.gbWithHandling <- function(gbkRaw, count=0) {
   gbkData <- tryCatch({

diff --git a/R/helpers.R b/R/helpers.R
@@ -1,7 +1,7 @@
 #!/usr/bin/env RScript
 #contributors=c("Gregory Smith", "Nils Jenke", "Michael Gruenstaeudl")
 #email="m_gruenstaeudl@fhsu.edu"
-#version="2024.03.22.1637"
+#version="2024.03.27.0311"
 
 HistCol <- function(cov, threshold, relative, logScale) {
   # Function to generate color vector for histogram data

diff --git a/R/parsingOps.R b/R/parsingOps.R
@@ -1,7 +1,7 @@
 #!/usr/bin/env RScript
 #contributors=c("Gregory Smith", "Nils Jenke", "Michael Gruenstaeudl")
 #email="m_gruenstaeudl@fhsu.edu"
-#version="2024.03.22.1637"
+#version="2024.03.27.0311"
 
 PACVr.parseGenes <- function (gbkSeqFeatures) {
   # Function to extract gene information from Genbank flatfile data