add fetchMSAmotif() function

.utilities.R

@@ -133,6 +133,50 @@ waitTimer <- function(t, nIntervals = 50) {
 }
 
 
+fetchMSAmotif <- function(ali, mot) {
+  # retrieve a subset from ali that spans the sequence in mot.
+  # Parameters:
+  #    ali        MsaAAMultipleAlignment object
+  #    mot  chr   substring within ali
+  # Value:  AAStringset
+
+  if (class(ali) != "MsaAAMultipleAlignment" &&
+      class(ali) != "MsaDNAMultipleAlignment" &&
+      class(ali) != "MsaRNAMultipleAlignment") {
+    stop("ali has to be an msa multiple alignment object.")
+  }
+
+  if (class(mot) != "character") {
+    stop("mot has to be a character object.")
+  }
+
+  x <- gsub("-", "", as.character(ali))  # pure sequence, no hyphens
+
+  idx <- grep(mot, x)[1] # first sequence containing mot. If no match,
+                         # idx becomes NA
+  if (is.na(idx)) {
+    stop("mot is not a subsequence in ali.")
+  }
+
+  # Find the match range
+  m <- regexpr(mot, x[idx])
+  motifStart <- as.numeric(m)
+  motifEnd <- attr(m, "match.length") + motifStart - 1
+
+  # Count characters, skip hyphens ...
+  x <- unlist(strsplit(as.character(ali)[idx], ""))
+  x <- x != "-"
+  x <- as.numeric(x)
+  x <- cumsum(x)
+
+  return(subseq(ali@unmasked,
+                start = which(x == motifStart)[1], # get the first position
+                end   = which(x == motifEnd)[1]))
+}
+
+
+
+
 # ====== PDB ID selection ======================================================
 
 selectPDBrep <- function(n) {

BIN-ALI-MSA.R

@@ -3,12 +3,13 @@
 # Purpose:  A Bioinformatics Course:
 #              R code accompanying the BIN-ALI-MSA unit.
 #
-# Version:  1.0
+# Version:  1.1
 #
-# Date:     2017  10  23
+# Date:     2017  10
 # Author:   Boris Steipe (boris.steipe@utoronto.ca)
 #
 # Versions:
+#           1.1    Added fetchMSAmotif()
 #           1.0    Fully refactored and rewritten for 2017
 #           0.1    First code copied from 2016 material.
 #
@@ -587,6 +588,12 @@ x <- cumsum(x)
 
 (motifAli <- subseq(msaM@unmasked, start = aliStart, end = aliEnd))
 
+# Packaging this into a function is convenient to have, therefore I have added
+# such a function to the .utilities.R script:  fetchMSAmotif(). Try it:
+
+wing <- "HEKVQGGFGKYQGTWV" # the MBP1_SACCE APSES "wing" sequence
+writeALN(fetchMSAmotif(msaM, wing))
+
 
 # ==   6.2  Plot a Sequence Logo  ==============================================