Outline

1
Outline
Outline

• Quiz 4
• Lab 1 (Quiz 3) Solution
• Scoping
• Algorithm efficiency
• Sorting
• Hashes
• Review for midterm
• Programming assignment 3

2
Lab 1 a More Elegant Tester
Lab1

• !/usr/bin/perl
• use strict
• use warnings
• File regex_tester.pl
• Author Jim Logan
• Date 6 October 2009
• Fully interactive version (i.e., no recompiles
  required) a regular expression
• tester based on a script by Fernando J. Pineda
  as presented to
• class of BINF623 by Jeff Solka on 10/5/09.
• Particularly useful in an Eclipse environment
  using its cut and paste facility.
• instructions for use
• print "\nAccepts keyboard entry of a regular
  expression and then permits\n"
• print "successive entry of strings to test that
  expression.\n"
• print "Square brackets in output indicate the
  text that matched pattern\n\n"
• print "Note Depending upon the environment
  (e.g. Eclipse), you may be\n"
• print "able to cut and paste into both the \"Next
  expression\" and the\n"

• initialization
• my regex '/./' default regex to
  start and to demonstrate
• my string 'This is a test string'
• my input ""
• my stripped_regex ""
• while (1) outer loop to sequence
  regular expressions
• print "\nCurrent regular expresssion
  regex\n"
• print "Enter a new expression to change or
  ENTER to continue without change.\n"
• print "(\"quit\" terminates the
  program)\n"
• print "New expression "
• input ltSTDINgt
• chomp input
• if (input /q./i) exit
• if (input ! //)
• regex input

3
Lab 1 a More Elegant Tester
Lab1

• stripped_regex substr (regex, 1, length
  (regex) -2)
• User includes the two slashes for a
  regular expresssion
• but they are stripped here so that
  variable is just the pattern
• that will be interpolated in /pattern/
  context.
• while (1) inner loop to sequence
  strings to test the expression
• print "\nCurrent test string
  string\n"
• print "Enter a new expression to
  change or ENTER to reset the regex.\n"
• print "New test string "
• input ltSTDINgt
• chomp input
• if (input //)
  for blank line, go back to set expresssion
• last

• else
• string input
  else run regex over input
• if( string /stripped_regex/)
  
• print("\'\n")
  show match in context of input
• else
• print("no match\n")
• exit never used

4
Lab 1 Solution
Lab1
Lab1

• What is a pattern that matches the substring
  world occurring
• anywhere in the input string, e.g.
• hello cold cruel world
• hello world news tonight
• helloworld.pl is a script
• Solution
• /world/

• 2. What is a pattern that matches the
• word world occurring anywhere in
• the input string, e.g.
• hello cold cruel world
• hello world news tonight
• but not
• helloworld.pl is a script
• Solution
• /\bworld\b/

5
Lab 1 Solution
Lab1

• 3. What is a pattern that matches the
• word world only if occurs at the end
• of the string, i.e
• hello cold cruel world
• but not
• next is world news tonight
• hello cold cruelworld
• Solution
• /\bworld\b/

• 4. What is a pattern that matches a
• string that starts with the word hello
• OR ends in the word world, e.g.
• hello and good night
• thats all for tonight world
• Solution
• /\bhello\b\bworld\b/

6
Lab 1 Solution
Lab1

• 5. What is a pattern that matches a
• string that starts with the word hello
• OR bye, AND ends with the word
• world, e.g.
• bye cold cruel world
• hello cold cruel world
• but not
• hello cold cruel world?
• hello cold cruelworld
• Solution
• /\b(hellobye)\b.\bworld\b/

• 6. What is a pattern that matches a
• substring world occurring 1 or more
• times at
• the end of the line, e.g.
• This string ends in world
• This string ends in worldworld
• This string ends in worldworldworld
• Solution
• /(world)/

7
Lab 1 Solution
Lab1

• 7. What is a pattern that matches one
• or more of backslashes immediately
• Followed by one or more asterisks, e.g.
• \\\\
• but not
• \\\\\
• Solution
• /\\\/

8
Lab 1 Solution
Lab1

• 8. What is a pattern that matches any line of
  input
• that has the same word repeated
• two or more times in a row. In this problem,
  words
• can be considered to be
• sequences of letters a to z, A to Z, digits, and
• underscores. Whitespace between
• words may differ, e.g.
• Paris in the the spring
• I thought that that was the problem
• For this example you will need to use
  backreferences. A
• backreference is a reference to a string captured
  with
• parentheses. (Recall that in Perl, captured
• strings are referred to as 1,,9) In a regular
  expression,
• you can refer to captured strings, while the
  pattern is being
• matched, as \1,\9. For example,
• /(AT)G(\1)/ matches a 5 character string ATGAT.
• Note Strictly speaking the inclusion of
  backreferences makes

• Solution
• /\b(\S)\b(\s\1\b)/
• Understanding this
• \b start at a word boundary (begin letters)
• (\S) find chunk of nonwhite space
• \b until another word boundary (end letters)
• (\s separated by some white space
• \1 and that very same chunk again
• \b) until another word boundary
• one or more sets of these

9
Be Careful With Scope
Scoping

• !/usr/bin/perl
• use strict
• use warnings
• my x 23
• print "value in main body is x \n"
• mysub(x)
• print "value in main body is x \n"
• exit
• sub mysub
• print "value in subroutine is x \n"
• x33

• !/usr/bin/perl
• use strict
• use warnings
• my x 23
• print "value in main body is x \n"
• mysub(x)
• print "value in main body is x \n"
• exit
• sub mysub
• print "value in subroutine is x \n"
• x33

10
Be Careful With Scope (cont.)
Scoping

• !/usr/bin/perl
• use strict
• use warnings
• my x 23
• print "value in main body is x \n"
• mysub(x)
• print "value in main body is x \n"
• exit
• sub mysub
• my(x) _at__
• x33
• print "value in subroutine is x \n"

• value in main body is 23
• value in subroutine is 33
• value in main body is 23

11
Data Structures and Algorithm Efficiency
Algorithm Efficiency
Algorithm is O(N2)

• An inefficient way to compute intersections
• my _at_a qw/ A B C D E F G H I J K X Y Z /
• my _at_b qw/ Q R S A C D T U G H V I J K X Z /
• my _at_intersection ()
• for my i (_at_a)
• for my j (_at_b)
• if (i eq j)
• push _at_intersection, i
• last
• print "_at_intersection\n"
• exit
• Output

N size of Lists
12
Data Structures and Algorithm Efficiency
Algorithm Efficiency

• A better way to compute intersections
• my _at_a qw/ A B C D E F G H I J K X Y Z /
• my _at_b qw/ Q R S A C D T U G H V I J K X Z /
• my _at_intersection ()
• "mark" each item in _at_a
• my mark ()
• for my i (_at_a) marki 1
• intersection any "marked" item in _at_b
• for my j (_at_b)
• if (exists markj)
• push _at_intersection, j
• print "_at_intersection\n"
• exit

version 1
version 2
13
Demonstration
Algorithm Efficiency

• Unix commands
• /usr/bin/time
• head
• diff
• cmp
• wc -l list1 list2
• 24762 list1
• 12381 list2
• 37143 total
• /usr/bin/time intersect1.pl list1 list2 gt out1
• 22.91 real 22.88 user 0.02
  sys
• /usr/bin/time intersect2.pl list1 list2 gt out2
• 0.06 real 0.05 user 0.00
  sys
• 22.88/.05 458

14
Hashes and Efficiency
Hashes

• Hashes provide a very fast way to look up
  information associated with a set of scalar
  values (keys)
• Examples
• Count how many time each word appears in a file
• Also whether or not a certain work appeared in a
  file
• Count how many time each codon appears in a DNA
  sequence
• Whether a given codon appears in a sequence
• How many time an item appears in a given list
• Intersections

15
Examples
Hashes

• Write a subroutine get_intersection(\_at_a, \_at_b)
  that returns the intersection of two lists.
• Write a subroutine first_list_only(\_at_a, \_at_b) that
  returns the items that are in list _at_a but not in
  _at_b.
• Write a subroutine unique(_at_a) that return the
  unique items in list _at_a (that is, remove the
  duplicates).
• Write a subroutine dups(n, _at_a) that returns a
  list of items that appear in _at_a at least n
  times.

16
Sorting
Sorting

• sort LIST -- returns list sorted in string order
• sort BLOCK LIST -- compares according to BLOCK
• sort USERSUB LIST -- compares according
  subroutine SUB

17
Sorting Our First Attempt
Sorting

• !/usr/bin/perl
• use strict
• use warnings
• my(_at_unsorted) (17, 8, 2, 111)
• my(_at_sorted) sort _at_unsorted
• print "_at_unsorted \n"
• print "_at_sorted \n"
• exit

• Output
• 17 8 2 111
• 111 17 2 8

18
The Comparison Operator
Sorting

• 1. a ltgt b returns 0 if equal, 1 if a gt b,
  -1 if a lt b
• 2. The "cmp" operator gives similar results for
  strings
• 3. a and b are special global variables
• do NOT declare with "my" and do NOT modify.

19
Sorting Numerically
Sorting

• !/usr/bin/perl
• use strict
• use warnings
• my(_at_unsorted) (17, 8, 2, 111)
• my(_at_sorted) sort a ltgt b _at_unsorted
• print "_at_unsorted \n"
• print "_at_sorted \n"
• exit

• Output
• 17 8 2 111
• 2 8 17 111

20
Sorting Using a Subroutine
Sorting

• !/usr/bin/perl
• use strict
• use warnings
• my(_at_unsorted) (17, 8, 2, 111)
• my(_at_sorted) sort numerically _at_unsorted
• print "_at_unsorted \n"
• print "_at_sorted \n"
• exit
• sub numerically a ltgt b

• Output
• 17 8 2 111
• 2 8 17 111

21
Sorting Descending
Sorting

• !/usr/bin/perl
• use strict
• use warnings
• my(_at_unsorted) (17, 8, 2, 111)
• my(_at_reversesorted) reverse sort numerically
  _at_unsorted
• print "_at_unsorted \n"
• print "_at_reversesorted \n"
• exit
• sub numerically a ltgt b

• Output
• 17 8 2 111
• 111 17 8 2

22
Sorting DNA by Length
Sorting

• !/usr/bin/perl
• use strict
• use warnings
• Sorting strings
• my _at_dna qw/ TATAATG TTTT GT CTCAT /
• Sort _at_dna by length
• _at_dna sort length(a) ltgt length(b) _at_dna
• print "_at_dna\n" Output GT TTTT CTCAT TATAATG
• exit

• Output
• GT TTTT CTCAT TATAATG

23
Sorting DNA by Number of Ts (Largest First)
Sorting

• !/usr/bin/perl
• use strict
• use warnings
• Sorting strings
• my _at_dna qw/ TATAATG TTTT GT CTCAT /
• _at_dna sort (b tr/Tt//) ltgt (a tr/Tt//)
  _at_dna
• print "_at_dna\n" Output TTTT TATAATG CTCAT GT
• exit

• Output
• TTTT TATAATG CTCAT GT

24
Sorting DNA by Number of Ts (Largest First)
(Take 2)
Sorting

• !/usr/bin/perl
• use strict
• use warnings
• Sorting strings
• my _at_dna qw/ TATAATG TTTT GT CTCAT /
• _at_dna reverse sort
• (a tr/Tt//) ltgt (b tr/Tt//) _at_dna
• print "_at_dna\n" Output TTTT TATAATG CTCAT GT
• exit

• Output
• TTTT TATAATG CTCAT GT

25
Sorting Strings Without Regard to Case
Sorting

• !/usr/bin/perl
• use strict
• use warnings
• Sort strings without regard to case
• my(_at_unsorted) qw/ mouse Rat HUMAN eColi /
• my(_at_sorted) sort lc(a) cmp lc(b)
  _at_unsorted
• print "_at_unsorted \n"
• print "_at_sorted \n"
• exit

• Output
• mouse Rat HUMAN eColi
• eColi HUMAN mouse Rat

26
Sorting Hashes by Value
Sorting

• !/usr/bin/perl
• use strict
• use warnings
• my(sales_amount) ( autogt100, kitchengt2000,
  hardwaregt200 )
• sub bysales sales_amountb ltgt
  sales_amounta
• for my dept (sort bysales keys sales_amount)
• printf "s\t4d\n", dept, sales_amountdept
• exit

• Output
• kitchen2000
• hardware 200
• auto 100

27
Review for Midterm BINF634
Midterm

• Material
• Tisdall Chapters 1-9
• Wall Chapter 5
• Lecture notes
• The exam will be open book and notes
• You cannot work together on it
• You cannot use outside material
• You will have the full period to take the midterm
• You will be asked to program

28
Some Example Questions
Midterm

• Given two DNA fragments contained in DNA1 and
  DNA2 how can we concatenate these to make a
  third string DNA3?

DNA3 DNA1DNA2

• DNA3 DNA1 . DNA2

29
Some Example Questions
Midterm

• What does this line of code do?
• RNA s/T/U/ig

Substitute Ts with Us in a case insensitive
manner globally within the string RNA
30
Some Example Questions
Midterm

• What does this statement do?
• revcom tr/ACGT/TGCA/

It performs the mapping A ? T C ? G G ? C T ?
A all at once
31
Some Example Questions
Midterm

• What do these four lines do?
• _at_bases (A, C, G, T)
• baset pop _at_bases
• unshift (_at_bases, base1)
• print _at_bases\n\n

T A G C
32
Some Example Questions
Midterm

• What does this code snippet do if COND is true
• unless(COND)
• do something

nothing
33
Some Example Questions
Midterm

• What does this code fragment do?
• protein join(,_at_protein)

Converts the array _at_protein into a scalar
protein with no space between The entries
34
Some Example Questions
Midterm

• What does this code fragment do?
• myfile myfile
• Open(MYFILE, gtmyfile)

Opens the file myfile with the file handle
MYFILE for writing
35
Some Example Questions
Midterm

• What does this code fragment do?
• while(DNA /a/ig)a

Counts the occurrences of the letter a or
A within the string DNA
36
Some Example Questions
Midterm

• What is the effect of using the command
• use strict
• at the beginning of your program?

It insists that your programs have all their
variables declared as my variables
37
Some Example Questions
Midterm

• What is contained in the reserved variable 0 and
  
• in the array _at_ARGV ?

0 contains the name of the program _at_ARGV
contains the command line arguments for the
program
38
Some Example Questions
Midterm

• What is the difference between pass by value
  and pass by reference ?

In pass by value you provide a subroutine with
a copy of your variable. In pass by reference
you provide a subroutine with a pointer to your
variable. In this manner the subroutine can
change the contents of the variable.
39
Some Example Questions
Midterm

• What is a pointer and what does it mean to
  dereference a pointer?

A pointer is an address in memory to a particular
variable. Dereferecing a pointer is the act of
obtaining the information that is stored at a
particular pointer location.
40
Some Example Questions
Midterm

• How do you invoke perl with the debugger?

perl - d
41
Some Example Questions
Midterm

• Given an array _at_verbs what is going on here?
• verbsrand _at_verbs

rand wants an integer so it uses scalar _at_verbs
rand then generates a random number between 0 and
length of the array _at_verbs. This is then
converted to an integer to index into _at_verbs