Note 9 egrep.txt

Class 9
# we can create and instantiate variables in a bash session
# here, we create the variable FILE and set its value to a string
# that is the absolute path to the file ‘american-english’
  205  FILE=/usr/share/dict/american-english
# this echo statement returns the value of the variable
  206  echo $FILE
# now let’s use it. cat-ing the dictionary file shows us that it’s just a file of
# words, one word to a line. that perfect for experimenting with egrep
  207  cat $FILE
# note that when we use egrep with string literals, most of the time we don’t
# have to put the string literal in quotes; however, we must put regular expressions
# in double-quotes
  208  egrep "cat" $FILE
  209  egrep "cat" $FILE | less
# now we anchor ‘cat’ to the beginning of the line
  210  egrep "^cat" $FILE
# now we anchor it to the end of the line
  211  egrep "cat$" $FILE
# the dot (.) is a metacharacter that matches to all characters except the
# newline character
  212  egrep "ca.t" $FILE
# parentheses are the grouping metacharacters. in this instance, we want all hits
# returned that match the pattern ‘ca’ appearing two or more times consecutively:
# eg, ‘caca’ or ‘cacaca’ or ‘cacacaca’, etc.
  213  egrep "(ca){2,}" $FILE
  214  egrep "$" $FILE
# i still haven’t figured out why this regex behaves the way it does
  215  egrep "\$" $FILE
# this is the regex for a blank line (newline inserted at the beginning of the line)
  216  egrep "^$" $FILE
# pattern to match: cat or kat
  217  egrep "(c|k)at" $FILE
  218  egrep "(c|k)at" $FILE | less
  219  egrep "ca+t" $FILE | less
# pattern to match: val, vaal, vaaal, etc.
  220  egrep "va+l" $FILE | less
  221  egrep "(va)+" $FILE | less
# pattern to match: na, nana, nanana, etc.
  222  egrep "(na)+" $FILE | less

# remember in class that a couple of these regexes threw me with their output
# a point i forgot to make in class is that running a regex against a dictionary file
# lets you test it to see if it matches expected matches.
  223  egrep "b*(na)+" $FILE | less
  224  egrep "b.+(na)+" $FILE | less
# here we finally matched ‘banana’. this regex was built to be very restrictive.
  225  egrep "^b.+(na){2,}" $FILE | less
  226  ls
# ok, let’s make a directory to hold our datafiles
  227  mkdir datafile

  229  cd datafile/
# notice that the second argument is the tilde (~), which is the shortcut
# for home directory
  230  mv reformatted_Unigene.fa ~
  231  cd
  232  ls
# remember that the mv command is also used to rename files

  233  mv reformatted_Unigene.fa datafile/
# now let’s set the value of the FILE variable to the relative path to our
# reformatted fasta file
  234  FILE=datafile/reformatted_Unigene.fa
# we’ll start with a literal string
  235  egrep "GATTACA" $FILE
# and now let’s see how many records match the pattern one or more times
# consecutively
  236  egrep "(GATTACA)+" $FILE | wc
# the -c option returns the count, as well
  237  egrep -c "(GATTACA)+" $FILE
  238  egrep "(GATTACA)+" $FILE
# return records that match the pattern two or more times consecutively
  239  egrep "(GATTACA){2,}" $FILE
# only one record contains GATTACAGATTACA
  240  egrep "(GATTACA){2,}" $FILE | wc
# agrep is another tool in the grep family. it allows approximate matches to patterns
# it matches strings that are exactly like the pattern, but also strings that vary
# from the pattern by insertions, deletions, and substitutions
# the default cost of an insertion, deletion, or substitution is 1.
# in the statement below, we allow for an approximate match of exactly one deletion,
# as we have set the ‘cost’ for insertions and substitutions too high
  242  agrep -1 -I2 -S2 GATTACAGATTACA $FILE
# we pipe the output to egrep to color matches to one part of the matching string
# this makes it easier for us to see the approximate matches in the
# nucleotide strings
  243  agrep -1 -I2 -S2 GATTACAGATTACA $FILE | egrep GATTACA
# here we allow for up to two deletions (insertions and substitutions are still too
# expensive
  244  agrep -2 -I3 -S3 GATTACAGATTACA $FILE | egrep GATTACA
# now we experiment with regular expressions against our fasta file
# note that some statements are functionally equivalent; that is, they yield the same
# result
  245  egrep "[AT]GATTACA" $FILE
  246  egrep "[^AT]GATTACA" $FILE
  247  egrep "[^ACGT]{6,}" $FILE
  248  egrep "[^ACGT]{16,}" $FILE
  249  egrep "(T|A)(A|T)GATTACA" $FILE
  250  egrep "(A|T){2,}GATTACA" $FILE