CSE 421 Algorithms Richard Anderson Lecture 19 Longest Common Subsequence.

Slides:



Advertisements
Similar presentations
Dynamic Programming Nithya Tarek. Dynamic Programming Dynamic programming solves problems by combining the solutions to sub problems. Paradigms: Divide.
Advertisements

Introduction to Bioinformatics Algorithms Divide & Conquer Algorithms.
Introduction to Bioinformatics Algorithms Divide & Conquer Algorithms.
Introduction to Bioinformatics Algorithms Divide & Conquer Algorithms.
CPSC 335 Dynamic Programming Dr. Marina Gavrilova Computer Science University of Calgary Canada.
CSEP 521 Applied Algorithms Richard Anderson Lecture 7 Dynamic Programming.
RAIK 283: Data Structures & Algorithms
Inexact Matching of Strings General Problem –Input Strings S and T –Questions How distant is S from T? How similar is S to T? Solution Technique –Dynamic.
Comp 122, Fall 2004 Dynamic Programming. dynprog - 2 Lin / Devi Comp 122, Spring 2004 Longest Common Subsequence  Problem: Given 2 sequences, X =  x.
1 Longest Common Subsequence (LCS) Problem: Given sequences x[1..m] and y[1..n], find a longest common subsequence of both. Example: x=ABCBDAB and y=BDCABA,
Space Efficient Alignment Algorithms and Affine Gap Penalties
Space Efficient Alignment Algorithms Dr. Nancy Warter-Perez June 24, 2005.
Dynamic Programming Optimization Problems Dynamic Programming Paradigm
CSE 421 Algorithms Richard Anderson Lecture 16 Dynamic Programming.
Dynamic Programming1. 2 Outline and Reading Matrix Chain-Product (§5.3.1) The General Technique (§5.3.2) 0-1 Knapsack Problem (§5.3.3)
UNIVERSITY OF SOUTH CAROLINA College of Engineering & Information Technology Bioinformatics Algorithms and Data Structures Chapter 11: Core String Edits.
CSE 421 Algorithms Richard Anderson Lecture 19 Longest Common Subsequence.
FA05CSE182 CSE 182-L2:Blast & variants I Dynamic Programming
Lecture 7 Topics Dynamic Programming
FA05CSE182 CSE 182-L2:Blast & variants I Dynamic Programming
Class 2: Basic Sequence Alignment
Space Efficient Alignment Algorithms Dr. Nancy Warter-Perez.
1 Theory I Algorithm Design and Analysis (11 - Edit distance and approximate string matching) Prof. Dr. Th. Ottmann.
Classroom Presenter and Tutored Video Instruction Richard Anderson Natalie Linnell University of Washington 1.
Dynamic Programming Adapted from Introduction and Algorithms by Kleinberg and Tardos.
Dynamic Programming Introduction to Algorithms Dynamic Programming CSE 680 Prof. Roger Crawfis.
Dynamic Programming – Part 2 Introduction to Algorithms Dynamic Programming – Part 2 CSE 680 Prof. Roger Crawfis.
Dynamic Programming (16.0/15) The 3-d Paradigm 1st = Divide and Conquer 2nd = Greedy Algorithm Dynamic Programming = metatechnique (not a particular algorithm)
Dynamic Programming Tutorial &Practice on Longest Common Sub-sequence.
Dynamic Programming.
6/4/ ITCS 6114 Dynamic programming Longest Common Subsequence.
Prof. Swarat Chaudhuri COMP 482: Design and Analysis of Algorithms Spring 2012 Lecture 16.
1 Chapter 6 Dynamic Programming. 2 Algorithmic Paradigms Greedy. Build up a solution incrementally, optimizing some local criterion. Divide-and-conquer.
Analysis of Algorithms CS 477/677 Instructor: Monica Nicolescu Lecture 14.
Sequence Alignment Tanya Berger-Wolf CS502: Algorithms in Computational Biology January 25, 2011.
Introduction to Bioinformatics Algorithms Divide & Conquer Algorithms.
Space Efficient Alignment Algorithms and Affine Gap Penalties Dr. Nancy Warter-Perez.
COSC 3101NJ. Elder Announcements Midterm Exam: Fri Feb 27 CSE C –Two Blocks: 16:00-17:30 17:30-19:00 –The exam will be 1.5 hours in length. –You can attend.
CS38 Introduction to Algorithms Lecture 10 May 1, 2014.
Part 2 # 68 Longest Common Subsequence T.H. Cormen et al., Introduction to Algorithms, MIT press, 3/e, 2009, pp Example: X=abadcda, Y=acbacadb.
Dynamic Programming Tutorial &Practice on Longest Common Sub-sequence.
TU/e Algorithms (2IL15) – Lecture 4 1 DYNAMIC PROGRAMMING II
Dynamic Programming Csc 487/687 Computing for Bioinformatics.
CS502: Algorithms in Computational Biology
Dynamic Programming Typically applied to optimization problems
Divide & Conquer Algorithms
Algorithmics - Lecture 11
Prepared by Chen & Po-Chuan 2016/03/29
Memory Efficient Longest Common Subsequence
Richard Anderson Lecture 20 LCS / Shortest Paths
Richard Anderson Lecture 13 Divide and Conquer
Richard Anderson Lecture 19 Longest Common Subsequence
Approximate Matching of Run-Length Compressed Strings
Dynamic Programming.
Lecture 8. Paradigm #6 Dynamic Programming
Trevor Brown DC 2338, Office hour M3-4pm
Bioinformatics Algorithms and Data Structures
Dynamic Programming.
Longest Common Subsequence
Richard Anderson Lecture 19 Memory Efficient Dynamic Programming
Richard Anderson Lecture 19 Longest Common Subsequence
Lecture 5 Dynamic Programming
Longest Common Subsequence
Richard Anderson Lecture 19 Memory Efficient Dynamic Programming
Richard Anderson Lecture 14 Divide and Conquer
Richard Anderson Lecture 18 Dynamic Programming
Dynamic Programming.
Richard Anderson Lecture 20 Space Efficient LCS
Memory Efficient Dynamic Programming / Shortest Paths
Presentation transcript:

CSE 421 Algorithms Richard Anderson Lecture 19 Longest Common Subsequence

C=c 1 …c g is a subsequence of A=a 1 …a m if C can be obtained by removing elements from A (but retaining order) LCS(A, B): A maximum length sequence that is a subsequence of both A and B ocurranec occurrence attacggct tacgacca

Determine the LCS of the following strings BARTHOLEMEWSIMPSON KRUSTYTHECLOWN

String Alignment Problem Align sequences with gaps Charge  x if character x is unmatched Charge  xy if character x is matched to character y CAT TGA AT CAGAT AGGA Note: the problem is often expressed as a minimization problem, with  xx = 0 and  x > 0

LCS Optimization A = a 1 a 2 …a m B = b 1 b 2 …b n Opt[ j, k] is the length of LCS(a 1 a 2 …a j, b 1 b 2 …b k )

Optimization recurrence If a j = b k, Opt[ j,k ] = 1 + Opt[ j-1, k-1 ] If a j != b k, Opt[ j,k] = max(Opt[ j-1,k], Opt[ j,k-1])

Give the Optimization Recurrence for the String Alignment Problem Charge  x if character x is unmatched Charge  xy if character x is matched to character y Opt[ j, k] = Let a j = x and b k = y Express as minimization

Dynamic Programming Computation

Code to compute Opt[j,k]

Storing the path information A[1..m], B[1..n] for i := 1 to m Opt[i, 0] := 0; for j := 1 to n Opt[0,j] := 0; Opt[0,0] := 0; for i := 1 to m for j := 1 to n if A[i] = B[j] { Opt[i,j] := 1 + Opt[i-1,j-1]; Best[i,j] := Diag; } else if Opt[i-1, j] >= Opt[i, j-1] { Opt[i, j] := Opt[i-1, j], Best[i,j] := Left; } else { Opt[i, j] := Opt[i, j-1], Best[i,j] := Down; } a 1 …a m b 1 …b n

How good is this algorithm? Is it feasible to compute the LCS of two strings of length 100,000 on a standard desktop PC? Why or why not.

Observations about the Algorithm The computation can be done in O(m+n) space if we only need one column of the Opt values or Best Values The algorithm can be run from either end of the strings

Algorithm Performance O(nm) time and O(nm) space On current desktop machines –n, m < 10,000 is easy –n, m > 1,000,000 is prohibitive Space is more likely to be the bounding resource than time

Observations about the Algorithm The computation can be done in O(m+n) space if we only need one column of the Opt values or Best Values The algorithm can be run from either end of the strings

Computing LCS in O(nm) time and O(n+m) space Divide and conquer algorithm Recomputing values used to save space

Divide and Conquer Algorithm Where does the best path cross the middle column? For a fixed i, and for each j, compute the LCS that has a i matched with b j

Constrained LCS LCS i,j (A,B): The LCS such that –a 1,…,a i paired with elements of b 1,…,b j –a i+1,…a m paired with elements of b j+1,…,b n LCS 4,3 (abbacbb, cbbaa)

A = RRSSRTTRTS B=RTSRRSTST Compute LCS 5,0 (A,B), LCS 5,1 (A,B),…,LCS 5,9 (A,B)

A = RRSSRTTRTS B=RTSRRSTST Compute LCS 5,0 (A,B), LCS 5,1 (A,B),…,LCS 5,9 (A,B) jleftright

Computing the middle column From the left, compute LCS(a 1 …a m/2,b 1 …b j ) From the right, compute LCS(a m/2+1 …a m,b j+1 …b n ) Add values for corresponding j’s Note – this is space efficient

Divide and Conquer A = a 1,…,a m B = b 1,…,b n Find j such that –LCS(a 1 …a m/2, b 1 …b j ) and –LCS(a m/2+1 …a m,b j+1 …b n ) yield optimal solution Recurse

Algorithm Analysis T(m,n) = T(m/2, j) + T(m/2, n-j) + cnm

Prove by induction that T(m,n) <= 2cmn

Memory Efficient LCS Summary We can afford O(nm) time, but we can’t afford O(nm) space If we only want to compute the length of the LCS, we can easily reduce space to O(n+m) Avoid storing the value by recomputing values –Divide and conquer used to reduce problem sizes