Basic Local Alignment Search Tool (BLAST)

1
Basic Local Alignment Search Tool (BLAST)

• Katie Moreland

2
Overview

• Sequence Alignment
• Dynamic Programming
• BLAST tutorial
• Example execution of BLAST
• References

3
Sequence Alignment

• In bioinformatics, a sequence alignment is a way
  of arranging the primary sequences of DNA, RNA,
  or protein to identify regions of similarity that
  may be a consequence of functional, structural,
  or evolutionary relationships between the
  sequences. (http//wikipedia.org)
• Example Alignment
• G A A T T C A G T T A
• G G A - T C - G - - A

4
Sequence Alignment Cont

• Motivations
• Similar primary structure in proteins implies
  similar form and function
• Similar short sequences can lead to motif finding
  (ie promoter regions)
• Similarities between gene regions can be used for
  phylogenetic classification

5
Sequence Similarity

• Alignments are not unique
• Need a way to compare alignments to find optimal
• Optimal Alignment is the alignment that maximizes
  the overall score (may not be unique)
• Three possibilities when aligning a character for
  each string (perfect match, mismatch, indel)
• Align the two characters
• Perfect Match Mismatch
• C C
• C G
• Insertion/Deletion (indel)
• Gap in 1st string (S) Gap in 2nd string (T)
• - C
• C -

6
Sequence Similarity Cont

• Simple Metric
• s(x,x) 1 (match)
• s(x,y) -1 (mismatch)
• s(x,-) s(-,x) -1 (indel)
• In practice it is useful to define a substitution
  matrix such as PAM250 to take probabilities of
  certain mutations into account.
• ie cost of mutation to a chemically similar
  amino-acid less than cost of mutation to
  dissimilar amino-acid
• Cost of indels depends on application

7
Intro to Dynamic Programming

• Used to reduce time complexity of algorithms with
  certain properties
• Characteristics of Dynamic Programming
• Overlapping subproblems (otherwise
  recursion/divide and conquer)
• Optimality of subproblems (ie Shortest Path)

8
Intro to Dynamic Programming

• Two types of alignment
• Global (Needleman-Wunsch)
• Attempt to align every residue in the sequences
• Most useful when sequences are similar in size
  and sequence
• Local (Smith-Waterman)
• Finds an alignment for parts of the two strings
• Most useful for dissimilar sequences that share
  regions of similarity or contain similar motifs

9
Needleman-Wunsch Algorithm

• Input Two strings, S and T
• Construct a matrix with S1 rows and T1
  columns
• Label each row with a symbol from S and each
  column with a symbol from T, except for the first
  position in each which represents an initial gap
• Beginning at upper left corner
• Move diagonally to represent aligning the two
  characters from the strings
• Move right to represent inserting a space in S
• Move down to represent insert a space in T
• Update when newScore gt oldScore (include arrow to
  show which cell we came from)
• Optimal alignment score is in bottom right corner
  of matrix
• Backtrack to find optimal alignment

10
Needleman-Wunsch Algorithm

• Sequences to Align
• S GCTC
• T CGTTC
• Simple Scoring Function
• s(x,x) 2 (match)
• s(x,y) -1 (mismatch)
• s(x,-) s(-,x) -1 (indel)

11
Tracing Needleman-Wunsch
C G T T C

G
C
T
C
12
Tracing Needleman-Wunsch
C G T T C
0
G
C
T
C
13
Tracing Needleman-Wunsch
C G T T C
0 -1
G -1 -1
C
T
C
14
Tracing Needleman-Wunsch
C G T T C
0 -1
G -1 -1
C
T
C
15
Tracing Needleman-Wunsch
C G T T C
0 -1 -2
G -1 -1 1
C
T
C
16
Tracing Needleman-Wunsch
C G T T C
0 -1 -2
G -1 -1 1
C
T
C
17
Tracing Needleman-Wunsch
C G T T C
0 -1 -2 -3
G -1 -1 1 -3
C
T
C
18
Tracing Needleman-Wunsch
C G T T C
0 -1 -2 -3
G -1 -1 1 -3
C
T
C
19
Tracing Needleman-Wunsch
C G T T C
0 -1 -2 -3 -4
G -1 -1 1 -3 -4
C
T
C
20
Tracing Needleman-Wunsch
C G T T C
0 -1 -2 -3 -4
G -1 -1 1 -3 -4
C
T
C
21
Tracing Needleman-Wunsch
C G T T C
0 -1 -2 -3 -4 -5
G -1 -1 1 -3 -4 -5
C
T
C
22
Tracing Needleman-Wunsch
C G T T C
0 -1 -2 -3 -4 -5
G -1 -1 1 -3 -4 -5
C
T
C
23
Tracing Needleman-Wunsch
C G T T C
0 -1 -2 -3 -4 -5
G -1 -1 1 -3 -4 -5
C
T
C
24
Tracing Needleman-Wunsch
C G T T C
0 -1 -2 -3 -4 -5
G -1 -1 1 -3 -4 -5
C
T
C
25
Tracing Needleman-Wunsch
C G T T C
0 -1 -2 -3 -4 -5
G -1 -1 1 -3 -4 -5
C -2 1
T
C
26
Tracing Needleman-Wunsch
C G T T C
0 -1 -2 -3 -4 -5
G -1 -1 1 -3 -4 -5
C -2 1
T
C
27
Tracing Needleman-Wunsch
C G T T C
0 -1 -2 -3 -4 -5
G -1 -1 1 -3 -4 -5
C -2 1 -2
T
C
28
Tracing Needleman-Wunsch
C G T T C
0 -1 -2 -3 -4 -5
G -1 -1 1 -3 -4 -5
C -2 1 -2
T
C
29
Tracing Needleman-Wunsch
C G T T C
0 -1 -2 -3 -4 -5
G -1 -1 1 0 -4 -5
C -2 1 0 0
T
C
30
Tracing Needleman-Wunsch
C G T T C
0 -1 -2 -3 -4 -5
G -1 -1 1 0 -4 -5
C -2 1 0 0
T
C
31
Tracing Needleman-Wunsch
C G T T C
0 -1 -2 -3 -4 -5
G -1 -1 1 0 -1 -5
C -2 1 0 0 -1
T
C
32
Tracing Needleman-Wunsch
C G T T C
0 -1 -2 -3 -4 -5
G -1 -1 1 0 -1 -5
C -2 1 0 0 -1
T
C
33
Tracing Needleman-Wunsch
C G T T C
0 -1 -2 -3 -4 -5
G -1 -1 1 0 -1 -2
C -2 1 0 0 -1 1
T
C
34
Tracing Needleman-Wunsch
C G T T C
0 -1 -2 -3 -4 -5
G -1 -1 1 0 -1 -2
C -2 1 0 0 -1 1
T
C
35
Tracing Needleman-Wunsch
C G T T C
0 -1 -2 -3 -4 -5
G -1 -1 1 0 -1 -2
C -2 1 0 0 -1 1
T
C
36
Tracing Needleman-Wunsch
C G T T C
0 -1 -2 -3 -4 -5
G -1 -1 1 0 -1 -2
C -2 1 0 0 -1 1
T
C
37
Tracing Needleman-Wunsch
C G T T C
0 -1 -2 -3 -4 -5
G -1 -1 1 0 -1 -2
C -2 1 0 0 -1 1
T -3 -3
C
38
Tracing Needleman-Wunsch
C G T T C
0 -1 -2 -3 -4 -5
G -1 -1 1 0 -1 -2
C -2 1 0 0 -1 1
T -3 -3
C
39
Tracing Needleman-Wunsch
C G T T C
0 -1 -2 -3 -4 -5
G -1 -1 1 0 -1 -2
C -2 1 0 0 -1 1
T -3 0 0
C
40
Tracing Needleman-Wunsch
C G T T C
0 -1 -2 -3 -4 -5
G -1 -1 1 0 -1 -2
C -2 1 0 0 -1 1
T -3 0 0
C
41
Tracing Needleman-Wunsch
C G T T C
0 -1 -2 -3 -4 -5
G -1 -1 1 0 -1 -2
C -2 1 0 0 -1 1
T -3 0 0 2
C
42
Tracing Needleman-Wunsch
C G T T C
0 -1 -2 -3 -4 -5
G -1 -1 1 0 -1 -2
C -2 1 0 0 -1 1
T -3 0 0 2
C
43
Tracing Needleman-Wunsch
C G T T C
0 -1 -2 -3 -4 -5
G -1 -1 1 0 -1 -2
C -2 1 0 0 -1 1
T -3 0 0 2 2
C
44
Tracing Needleman-Wunsch
C G T T C
0 -1 -2 -3 -4 -5
G -1 -1 1 0 -1 -2
C -2 1 0 0 -1 1
T -3 0 0 2 2
C
45
Tracing Needleman-Wunsch
C G T T C
0 -1 -2 -3 -4 -5
G -1 -1 1 0 -1 -2
C -2 1 0 0 -1 1
T -3 0 0 2 2 -2
C
46
Tracing Needleman-Wunsch
C G T T C
0 -1 -2 -3 -4 -5
G -1 -1 1 0 -1 -2
C -2 1 0 0 -1 1
T -3 0 0 2 2 -2
C
47
Tracing Needleman-Wunsch
C G T T C
0 -1 -2 -3 -4 -5
G -1 -1 1 0 -1 -2
C -2 1 0 0 -1 1
T -3 0 0 2 2 0
C
48
Tracing Needleman-Wunsch
C G T T C
0 -1 -2 -3 -4 -5
G -1 -1 1 0 -1 -2
C -2 1 0 0 -1 1
T -3 0 0 2 2 0
C
49
Tracing Needleman-Wunsch
C G T T C
0 -1 -2 -3 -4 -5
G -1 -1 1 0 -1 -2
C -2 1 0 0 -1 1
T -3 0 0 2 2 0
C -4 -1
50
Tracing Needleman-Wunsch
C G T T C
0 -1 -2 -3 -4 -5
G -1 -1 1 0 -1 -2
C -2 1 0 0 -1 1
T -3 0 0 2 2 0
C -4 -1
51
Tracing Needleman-Wunsch
C G T T C
0 -1 -2 -3 -4 -5
G -1 -1 1 0 -1 -2
C -2 1 0 0 -1 1
T -3 0 0 2 2 0
C -4 -1 -1
52
Tracing Needleman-Wunsch
C G T T C
0 -1 -2 -3 -4 -5
G -1 -1 1 0 -1 -2
C -2 1 0 0 -1 1
T -3 0 0 2 2 0
C -4 -1 -1
53
Tracing Needleman-Wunsch
C G T T C
0 -1 -2 -3 -4 -5
G -1 -1 1 0 -1 -2
C -2 1 0 0 -1 1
T -3 0 0 2 2 0
C -4 -1 -1 -1
54
Tracing Needleman-Wunsch
C G T T C
0 -1 -2 -3 -4 -5
G -1 -1 1 0 -1 -2
C -2 1 0 0 -1 1
T -3 0 0 2 2 0
C -4 -1 -1 -1
55
Tracing Needleman-Wunsch
C G T T C
0 -1 -2 -3 -4 -5
G -1 -1 1 0 -1 -2
C -2 1 0 0 -1 1
T -3 0 0 2 2 0
C -4 -1 -1 1 1
56
Tracing Needleman-Wunsch
C G T T C
0 -1 -2 -3 -4 -5
G -1 -1 1 0 -1 -2
C -2 1 0 0 -1 1
T -3 0 0 2 2 0
C -4 -1 -1 1 1
57
Tracing Needleman-Wunsch
C G T T C
0 -1 -2 -3 -4 -5
G -1 -1 1 0 -1 -2
C -2 1 0 0 -1 1
T -3 0 0 2 2 1
C -4 -1 -1 1 1 4
58
Tracing Needleman-Wunsch
C G T T C
0 -1 -2 -3 -4 -5
G -1 -1 1 0 -1 -2
C -2 1 0 0 -1 1
T -3 0 0 2 2 1
C -4 -1 -1 1 1 4
59
Tracing Needleman-Wunsch
C G T T C
0 -1 -2 -3 -4 -5
G -1 -1 1 0 -1 -2
C -2 1 0 0 -1 1
T -3 0 0 2 2 1
C -4 -1 -1 1 1 4
60
Tracing Needleman-Wunsch
C G T T C
0 -1 -2 -3 -4 -5
G -1 -1 1 0 -1 -2
C -2 1 0 0 -1 1
T -3 0 0 2 2 1
C -4 -1 -1 1 1 4
61
Tracing Needleman-Wunsch
C G T T C
0 -1 -2 -3 -4 -5
G -1 -1 1 0 -1 -2
C -2 1 0 0 -1 1
T -3 0 0 2 2 1
C -4 -1 -1 1 1 4
62
Tracing Needleman-Wunsch
C G T T C
0 -1 -2 -3 -4 -5
G -1 -1 1 0 -1 -2
C -2 1 0 0 -1 1
T -3 0 0 2 2 1
C -4 -1 -1 1 1 4
63
Tracing Needleman-Wunsch
C G T T C
0 -1 -2 -3 -4 -5
G -1 -1 1 0 -1 -2
C -2 1 0 0 -1 1
T -3 0 0 2 2 1
C -4 -1 -1 1 1 4
64
Tracing Needleman-Wunsch
C G T T C
0 -1 -2 -3 -4 -5
G -1 -1 1 0 -1 -2
C -2 1 0 0 -1 1
T -3 0 0 2 2 1
C -4 -1 -1 1 1 4
65
Tracing Needleman-Wunsch
C G T T C
0 -1 -2 -3 -4 -5
G -1 -1 1 0 -1 -2
C -2 1 0 0 -1 1
T -3 0 0 2 2 1
C -4 -1 -1 1 1 4
66
Tracing Needleman-Wunsch
C G T T C
0 -1 -2 -3 -4 -5
G -1 -1 1 0 -1 -2
C -2 1 0 0 -1 1
T -3 0 0 2 2 1
C -4 -1 -1 1 1 4
67
Tracing Needleman-Wunsch
C G T T C
0 -1 -2 -3 -4 -5
G -1 -1 1 0 -1 -2
C -2 1 0 0 -1 1
T -3 0 0 2 2 1
C -4 -1 -1 1 1 4
68
Tracing Needleman-Wunsch
C G T T C
0 -1 -2 -3 -4 -5
G -1 -1 1 0 -1 -2
C -2 1 0 0 -1 1
T -3 0 0 2 2 1
C -4 -1 -1 1 1 4
69
Tracing Needleman-Wunsch
C G T T C
0 -1 -2 -3 -4 -5
G -1 -1 1 0 -1 -2
C -2 1 0 0 -1 1
T -3 0 0 2 2 1
C -4 -1 -1 1 1 4
70
Tracing Needleman-Wunsch
C G T T C
0 -1 -2 -3 -4 -5
G -1 -1 1 0 -1 -2
C -2 1 0 0 -1 1
T -3 0 0 2 2 1
C -4 -1 -1 1 1 4
71
Tracing Needleman-Wunsch
C G T T C
0 -1 -2 -3 -4 -5
G -1 -1 1 0 -1 -2
C -2 1 0 0 -1 1
T -3 0 0 2 2 1
C -4 -1 -1 1 1 4
72
Tracing Needleman-Wunsch
C G T T C
0 -1 -2 -3 -4 -5
G -1 -1 1 0 -1 -2
C -2 1 0 0 -1 1
T -3 0 0 2 2 1
C -4 -1 -1 1 1 4
73
Tracing Needleman-Wunsch
C G T T C
0 -1 -2 -3 -4 -5
G -1 -1 1 0 -1 -2
C -2 1 0 0 -1 1
T -3 0 0 2 2 1
C -4 -1 -1 1 1 4
74
Tracing Needleman-Wunsch
C G T T C
0 -1 -2 -3 -4 -5
G -1 -1 1 0 -1 -2
C -2 1 0 0 -1 1
T -3 0 0 2 2 1
C -4 -1 -1 1 1 4
75
Tracing Needleman-Wunsch
C G T T C
0 -1 -2 -3 -4 -5
G -1 -1 1 0 -1 -2
C -2 1 0 0 -1 1
T -3 0 0 2 2 1
C -4 -1 -1 1 1 4
76
Tracing Needleman-Wunsch
C G T T C
0 -1 -2 -3 -4 -5
G -1 -1 1 0 -1 -2
C -2 1 0 0 -1 1
T -3 0 0 2 2 1
C -4 -1 -1 1 1 4
77
Tracing Needleman-Wunsch
C G T T C
0 -1 -2 -3 -4 -5
G -1 -1 1 0 -1 -2
C -2 1 0 0 -1 1
T -3 0 0 2 2 1
C -4 -1 -1 1 1 4
78
Modifications for Local Alignment

• Allow the algorithm to restart whenever it is
  advantageous to do so (start the algorithm from
  any position in S or T)
• If 0 gt newScore, set score for cell I,j to 0
• The optimal score is now the maximum value in all
  cells of the matrix (stop at any position in S or
  T)

79
Other Modifications

• Use a gap penalty function to accommodate large
  areas of gaps vs many gaps of size 1
• Biological motivations (ie mutations, cDNA
  matching)

80
BLAST

• Basic Local Alignment Search Tool
• http//www.ncbi.nlm.nih.gov/BLAST/
• Features
• Finds regions of local similarity between
  sequences
• Heuristic approach achieves efficiency
  (important when searching entire databases of
  sequences)
• Computes statistical significance of matches
• Uses
• Infer evolutionary/functional relationships
• Identify members of gene families

81
BLAST Algorithm

• Three Stages
• Find hotspots exact matches of word lengthW in
  the two sequences being considered (idea good
  alignments for sequences will share regions of
  similarity, find first)
• Extend hotspots in both directions using ungapped
  alignment to increase alignment score, pass high
  scoring sequences to stage 3
• Perform gapped alignment between the 2 sequences
  using variation of Smith-Waterman algorithm. Only
  statistically significant alignments are
  displayed to the user.

82
BLAST Input

• FASTA format
• gtgi532319pirTVFV2ETVFV2E envelope protein
  ELRLRYCAPAGFALLKCNDADYDGFKTNCSNVSVVHCTNLM NTTVTTG
  LLLNGSYSENRTQIWQKHRTSNDSALILLNKHYNL TVTCKRPGNKTVLP
  VTIMAGLVFHSQKYNLRLRQAWCHFPS NWKGAWKEVKEEIVNLPKERYR
  GTNDPKRIFFQRQWGDPE TANLWFNCHGEFFYCKMDWFLNYLNNLTVDA
  DHNECKNTS GTKSGNKRAPGPCVQRTYVACHIRSVIIWLETISKKTYAP
  PRE GHLECTSTVTGMTVELNYIPKNRTNVTLSPQIESIWAAELDRY KL
  VEITPIGFAPTEVRRYTGGHERQKRVPFVXXXXXXXXXXX XXXXXXXXX
  XXVQSQHLLAGILQQQKNL LAAVEAQQQMLKLTIWGVK
• Accession/GI Number
• Found using GenBank
• In FASTA example, gi number is 532319

83
BLAST Input
84
BLAST Options

• Select Program
• blastp, blastn, etc
• Select database(s) to search
• Nr default, contains GenBank, PDB, SwissProt, and
  others
• Gapped/Ungapped Alignment
• Search within certain organism

85
BLAST Options Cont

• Filtering on/off
• On by default, locates low complexity regions in
  a sequence and removes them before performing an
  alignment
• Low complexity region a region with highly
  biased amino acid composition
• E Value Threshold
• Default 10, represents the number of hits one
  can expect to find by chance when searching the
  database
• Substitution Matrix
• Default BLOSUM62
• Assigns probability for each alignment position
  that a given substitution is known to occur
• Other matrices are supported, including PAM
  matrices

86
BLAST Options
87
Advanced BLAST Options

• -G Cost to open a gap Integer
• default 11
• -E Cost to extend a gap Integer
• default 1
• -e Expectation value (E) Real
• default 10.0
• -W Word size
• default is 11 for blastn, 3 for other programs.
• -v Number of one-line descriptions (V) Integer
• default 100
• -b Number of alignments to show (B) Integer
• default 100

88
BLAST Output

• Request ID
• Query Information
• Database Information
• Taxonomy Reports Link
• Graphical Display of alignments
• Description of significant alignments
• Pairwise alignments

89
BLAST Output Cont
90
Taxonomy Reports

• Lineage Report
• Hierarchical tree structure representing how many
  hits occurred in each group
• 'focused' on the organism which yielded the
  strongest BLAST hit
• Organism Report
• Groups hits by species
• Taxonomy Report
• Summary of relationships between organisms in
  BLAST hit list

91
Graphical Display of Alignments

• displays the top 100 sequence alignments for a
  search by default
• Thick red bar at top represents query sequence,
  numbers correspond to amino acid residues
• Hits represented by colored bars, mouse over the
  bar to view the definition and score in the text
  box, click to go to pairwise alignment
• Bar color represents alignment similarity score
• Color Key given above query sequence to determine
  ranges of similarities for a particular color

92
Graphical Display of Alignments
93
Description of Significant Alignments

• Listed in order of decreasing significance
• Default number displayed100

94
Pairwise Alignments
95
BLAST Demonstration

• gtgi2501594spQ57997Y577_METJA PROTEIN MJ0577
• MSVMYKKILYPTDFSETAEIALKHVKAFKTLKAEEVILLHVIDEREIKK
  RDIFSLLLGVAGLNKSVEEFENELKNKLTEEAKNKMENIKKELEDVGFKV
  KDIIVVGIPHEEIVKIAEDEGVDIIIMGSHGKTNLKEILLGSVTENVIKK
  SNKPVLVVKRKNS
• http//www.ncbi.nlm.nih.gov/BLAST/

96
References

• Altschul, SF, W Gish, W Miller, EW Myers, and DJ
  Lipman. Basic local alignment search tool. J Mol
  Biol 215(3)403-10, 1990."
• 2. BLAST Tutorials
• http//www.ncbi.nlm.nih.gov/Education/BLASTinfo/i
  nformation3.html
• http//www.ornl.gov/sci/techresources/Human_Genom
  e/posters/chromosome/blast.shtml
• http//wikipedia.org
• 4. Hatzivassiloglou, V. http//www.hlt.utdalla
  s.edu/7Evh/Courses/Fall06/Lectures/Alignment20pa
  rt203.ppt