Exon prediction by Genomic Sequence alignment Burkhard Morgenstern and Oliver Rinner

1

Burkhard Morgenstern Institut für Mikrobiologie
und Genetik Molekulare Evolution und
Rekonstruktion von phylogenetischen Bäumen WS
2006/2007


2

• Goal
• Phylogeny reconstruction based on molecular
  sequence data (DNA, RNA, protein sequences)

3
Multiple sequence alignment

• Molecular phylogeny reconstruction relies on
  comparative nucleic acid and protein sequence
  analysis
• Alignment most important tool for sequence
  comparison
• Multiple alignment contains more information than
  pair-wise alignment

4
Tools for multiple sequence alignment

• Y I M Q E V Q Q E R
• Sequence duplicates in history (e.g. speciation
  event)

5
Tools for multiple sequence alignment

• Y I M Q E V Q Q E R

6
Tools for multiple sequence alignment

• Y I M Q E V Q Q E R
• Y I M Q E V Q Q E R

7
Tools for multiple sequence alignment

• Y I M Q E A Q Q E R
• Y L M Q E V Q Q E R
• Substitutions occur

8
Tools for multiple sequence alignment

• Y I M Q E A Q Q E R
• Y L M Q E V Q Q E R

9
Tools for multiple sequence alignment

• YAI M Q E A Q Q E R
• Y L M - - V Q Q E R V
• Insertions/deletions (indels) occur

10
Tools for multiple sequence alignment

• YAI M Q E A Q Q E R
• Y L M - - V Q Q E R V

11
Tools for multiple sequence alignment

• Y A I M Q E A Q Q E R
• Y L M V Q Q E R V
• because of insertions/deletions sequence
  similarity no longer immediately visible!

12
Tools for multiple sequence alignment

• Y A I M Q E A Q Q E R -
• Y - L M V - - Q Q E R V
• Alignment brings together related parts of the
  sequences by inserting gaps into sequences

13
Tools for multiple sequence alignment

• Y A I M Q E A Q Q E R -
• Y - L M V - - Q Q E R V

14
Tools for multiple sequence alignment

• Y A I M Q E A Q Q E R -
• Y - L M V - - Q Q E R V
• Mismatches correspond to substitutions
• Gaps correspond to indels

15
Tools for multiple sequence alignment

• Pairwise alignment alignment of two sequences
• Multiple alignment alignment of N gt 2 sequences

16
Tools for multiple sequence alignment

• s1 R Y I M R E A Q Y E S A Q
• s2 R C I V M R E A Y E
• s3 Y I M Q E V Q Q E R
• s4 W R Y I A M R E Q Y E
• Assumtion sequence family related by common
  ancestry similarity due to common history
• Sequence similarity not obvious (insertions and
  deletions may have happened)

17
Tools for multiple sequence alignment

• s1 - R Y I - M R E A Q Y E S A Q
• s2 - R C I V M R E A - Y E - - -
• s3 - - Y I - M Q E V Q Q E R - -
• s4 W R Y I A M R E - Q Y E - - -
• Multiple alignment arrangement of sequences by
  introducing gaps
• Alignment reveals sequence similarities

18
Tools for multiple sequence alignment

• s1 - R Y I - M R E A Q Y E S A Q
• s2 - R C I V M R E A - Y E - - -
• s3 - - Y I - M Q E V Q Q E R - -
• s4 W R Y I A M R E - Q Y E - - -

19
Tools for multiple sequence alignment

• s1 - R Y I - M R E A Q Y E S A Q
• s2 - R C I V M R E A - Y E - - -
• s3 - - Y I - M Q E V Q Q E R - -
• s4 W R Y I A M R E - Q Y E - - -

20
Tools for multiple sequence alignment

• s1 - R Y I - M R E A Q Y E S A Q
• s2 - R C I V M R E A - Y E - - -
• s3 - - Y I - M Q E V Q Q E R - -
• s4 W R Y I A M R E - Q Y E - - -
• General information in multiple alignment
• Functionally important regions more conserved
  than non-functional regions
• Local sequence conservation indicates
  functionality!

21
Tools for multiple sequence alignment

• s1 - R Y I - M R E A Q Y E S A Q
• s2 - R C I V M R E A - Y E - - -
• s3 - - Y I - M Q E V Q Q E R - -
• s4 W R Y I A M R E - Q Y E - - -
• Phylogeny reconstruction based on multiple
  alignment
• Estimate pairwise distances between sequences
  (distance-based methods for tree reconstruction)
• Estimate evloutionary events in evolution
  (parsimony and maximum likelihood methods)

22
Tools for multiple sequence alignment

• s1 - R Y I - M R E A Q Y E S A Q
• s2 - R C I V M R E A - Y E - - -
• s3 - - Y I - M Q E V Q Q E R - -
• s4 W R Y I A M R E - Q Y E - - -
• Task in bioinformatics Find best multiple
  alignment for given sequence set

23
Tools for multiple sequence alignment

• s1 - R Y I - M R E A Q Y E S A Q
• s2 - R C I V M R E A - Y E - - -
• s3 - - Y I - M Q E V Q Q E R - -
• s4 W R Y I A M R E - Q Y E - - -
• Astronomical number of possible alignments!

24
Tools for multiple sequence alignment

• s1 - R Y I - M R E A Q Y E S A Q
• s2 - R C I V M R E A - - - Y E -
• s3 Y I - - - M Q E V Q Q E R - -
• s4 W R Y I A M R E - Q Y E - - -
• Astronomical number of possible alignments!

25
Tools for multiple sequence alignment

• s1 - R Y I - M R E A Q Y E S A Q
• s2 - R C I V M R E A - - - Y E -
• s3 Y I - - - M Q E V Q Q E R - -
• s4 W R Y I A M R E - Q Y E - - -
• Computer has to decide which one is best??

26
Tools for multiple sequence alignment

• Questions in development of alignment programs
• (1) What is a good alignment?
• ? objective function (score)
• (2) How to find a good alignment?
• ? optimization algorithm
• First question far more important !

27
Tools for multiple sequence alignment

• Before defining an objective function (scoring
  scheme)
• What is a biologically good alignment ??

28
Tools for multiple sequence alignment

• Criteria for alignment quality
• 3D-Structure align residues at corresponding
  positions in 3D structure of protein!

29
Tools for multiple sequence alignment

• Criteria for alignment quality

30
Tools for multiple sequence alignment

• Criteria for alignment quality
• 3D-Structure align residues at corresponding
  positions in 3D structure of protein!

31
Tools for multiple sequence alignment

• Species related by common history

32
Tools for multiple sequence alignment

• Genes / proteins related by common history

33
Tools for multiple sequence alignment

• Criteria for alignment quality
• 3D-Structure align residues at corresponding
  positions in 3D structure of protein!
• Evolution align residues with common ancestors!

34
Tools for multiple sequence alignment

• s1 - R Y I - M R E A Q Y E S A Q
• s2 - R C I V M R E A - Y E - - -
• s3 - - Y I - M Q E V Q Q E R - -
• s4 W R Y I A M R E - Q Y E - - -
• Alignment hypothesis about sequence evolution
• Mismatches correspond to substitutions
• Gaps correspond to insertions/deletions

35
Tools for multiple sequence alignment

• s1 - R Y I - M R E A Q Y E S A Q
• s2 - R C I V M R E A - Y E - - -
• s3 - - Y I - M Q E V Q Q E R - -
• s4 W R Y I A M R E - Q Y E - - -
• Alignment hypothesis about sequence evolution
• Search for most plausible scenario!
• Estimate probabilities for individual
  evolutionary events insertions/deletions,
  substitutions

36
Tools for multiple sequence alignment

• s1 - R Y I - M R E A Q Y E S A Q
• s2 - R C I V M R E A - Y E - - -
• s3 - Y - I - M Q E V Q Q E R - -
• s4 W R Y I A M R E - Q Y E - - -
• Alignment hypothesis about sequence evolution
• Search for most plausible scenario!
• Estimate probabilities for individual
  evolutionary events insertions/deletions,
  substitutions

37
Tools for multiple sequence alignment

• Compute score s(a,b) for degree of similarity
  between amino acids a and b based on probability
• pa,b
• of substitution
• a ? b (or b ? a)
• (Extremely simplified!)

38
Tools for multiple sequence alignment
39
Tools for multiple sequence alignment

• Reason for different substitutin probabilities
  pa,b
• Different physical and chemical properties of
  amino acids
• Amino acids with similar properties more likely
  to be substituted against each other

40
(No Transcript)
41
Tools for multiple sequence alignment

• Use penalty for gaps introduced into alignment
• Simplest approach linear gap costs penalty
  proportional to gap length
• Non-linear gap penalties more realistic long gap
  caused by single insertion/deletion
• Most frequently used affine linear gap
  penalties more realistic, but efficient to
  calculate!

42

• Traditional Objective functions
• Define Score of alignments as
• Sum of individual similarity scores s(a,b)
• Minus gap penalties
• Needleman-Wunsch scoring system for pairwise
  alignment (1970)

43
Pair-wise sequence alignment

• T Y W I V
• T - - L V
• Example
• Score s(T,T) s(I,L) s (V,V) 2 g
• Assumption linear gap penalty!

44
Pair-wise sequence alignment

• T Y W I V
• T - - L V
• Dynamic-programming algorithm finds
• alignment with best score.
• (Needleman and Wunsch, 1970)

45
Pair-wise sequence alignment

• T Y W I V
• T - - L V
• Running time proportional to product of sequence
  length
• Time-complexity O(l1 l2)

46
Pair-wise sequence alignment

• Algorithm for pairwise alignment can be
  generalized to multiple alignment of N sequences
• Time-complexity O(l1 l2 lN)
• Not feasable in reality (too long running time!)
• Heuristic necessary, i.e. fast algorithm that
  does not necessarily produce mathematically best
  alignment

47
Progressive Alignment

• Most popular approach to (global) multiple
  sequence alignment
• Progressive Alignment
• Since mid-Eighties Feng/Doolittle,
  Higgins/Sharp, Taylor,

48
Progressive Alignment

• WCEAQTKNGQGWVPSNYITPVN
• WWRLNDKEGYVPRNLLGLYP
• AVVIQDNSDIKVVPKAKIIRD
• YAVESEAHPGSFQPVAALERIN
• WLNYNETTGERGDFPGTYVEYIGRKKISP

49
Progressive Alignment

• WCEAQTKNGQGWVPSNYITPVN
• WWRLNDKEGYVPRNLLGLYP
• AVVIQDNSDIKVVPKAKIIRD
• YAVESEAHPGSFQPVAALERIN
• WLNYNETTGERGDFPGTYVEYIGRKKISP
• Guide tree

50
Progressive Alignment

• WCEAQTKNGQGWVPSNYITPVN
• WW--RLNDKEGYVPRNLLGLYP-
• AVVIQDNSDIKVVP--KAKIIRD
• YAVESEASFQPVAALERIN
• WLNYNEERGDFPGTYVEYIGRKKISP
• Profile alignment, once a gap - always a gap

51
Progressive Alignment

• WCEAQTKNGQGWVPSNYITPVN
• WW--RLNDKEGYVPRNLLGLYP-
• AVVIQDNSDIKVVP--KAKIIRD
• YAVESEASVQ--PVAALERIN------
• WLN-YNEERGDFPGTYVEYIGRKKISP
• Profile alignment, once a gap - always a gap

52
Progressive Alignment

• WCEAQTKNGQGWVPSNYITPVN-
• WW--RLNDKEGYVPRNLLGLYP-
• AVVIQDNSDIKVVP--KAKIIRD
• YAVESEASVQ--PVAALERIN------
• WLN-YNEERGDFPGTYVEYIGRKKISP
• Profile alignment, once a gap - always a gap

53
Progressive Alignment

• WCEAQTKNGQGWVPSNYITPVN--------
• WW--RLNDKEGYVPRNLLGLYP--------
• AVVIQDNSDIKVVP--KAKIIRD-------
• YAVESEA---SVQ--PVAALERIN------
• WLN-YNE---ERGDFPGTYVEYIGRKKISP
• Profile alignment, once a gap - always a gap

54
Progressive Alignment

• WCEAQTKNGQGWVPSNYITPVN--------
• WW--RLNDKEGYVPRNLLGLYP--------
• AVVIQDNSDIKVVP--KAKIIRD-------
• YAVESEA---SVQ--PVAALERIN------
• WLN-YNE---ERGDFPGTYVEYIGRKKISP
• Most important implementation CLUSTAL W

55
Progressive Alignment

• CLUSTAL W Thompson et al., 1994 (17.000
  citations)
• Pairwise distances as 1 - percentage of identity
• Calculate un-rooted tree with Neighbor Joining
• Define root as central position in tree
• Define sequence weights based on tree
• Gap penalties calculated based on various
  parameters

56
Tools for multiple sequence alignment

• Problems with traditional approach
• Results depend on gap penalty
• Heuristic guide tree determines alignment
  alignment used for phylogeny reconstruction
• Algorithm produces global alignments.

57
Tools for multiple sequence alignment

• Problems with traditional approach
• But
• Many sequence families share only local
  similarity
• E.g. sequences share one conserved motif

58
The DIALIGN approach

• Morgenstern, Dress, Werner (1996),
• PNAS 93, 12098-12103
• Combination of global and local methods
• Assemble multiple alignment from
• gap-free local pair-wise alignments
• (,,fragments)

59
The DIALIGN approach

• atctaatagttaaactcccccgtgcttag
• cagtgcgtgtattactaacggttcaatcgcg
• caaagagtatcacccctgaattgaataa

60
The DIALIGN approach

• atctaatagttaaactcccccgtgcttag
• cagtgcgtgtattactaacggttcaatcgcg
• caaagagtatcacccctgaattgaataa

61
The DIALIGN approach

• atctaatagttaaactcccccgtgcttag
• cagtgcgtgtattactaacggttcaatcgcg
• caaagagtatcacccctgaattgaataa

62
The DIALIGN approach

• atctaatagttaaactcccccgtgcttag
• cagtgcgtgtattactaacggttcaatcgcg
• caaagagtatcacccctgaattgaataa

63
The DIALIGN approach

• atctaatagttaaactcccccgtgcttag
• cagtgcgtgtattactaacggttcaatcgcg
• caaagagtatcacccctgaattgaataa

64
The DIALIGN approach

• atctaatagttaaactcccccgtgcttag
• cagtgcgtgtattactaacggttcaatcgcg
• caaagagtatcacccctgaattgaataa

65
The DIALIGN approach

• atc------taatagttaaactcccccgtgcttag
• cagtgcgtgtattactaacggttcaatcgcg
• caaagagtatcacccctgaattgaataa

66
The DIALIGN approach

• atc------taatagttaaactcccccgtgcttag
• cagtgcgtgtattactaacggttcaatcgcg
• caaa--gagtatcacccctgaattgaataa

67
The DIALIGN approach

• atc------taatagttaaactcccccgtgcttag
• cagtgcgtgtattactaacggttcaatcgcg
• caaa--gagtatcacc----------cctgaattgaataa

68
The DIALIGN approach

• atc------taatagttaaactcccccgtgc-ttag
• cagtgcgtgtattactaac----------gg-ttcaatcgcg
• caaa--gagtatcacc----------cctgaattgaataa

69
The DIALIGN approach
Consistency!

• atc------taatagttaaactcccccgtgc-ttag
• cagtgcgtgtattactaac----------gg-ttcaatcgcg
• caaa--gagtatcacc----------cctgaattgaataa

70
The DIALIGN approach

• atc------TAATAGTTAaactccccCGTGC-TTag
• cagtgcGTGTATTACTAAc----------GG-TTCAATcgcg
• caaa--GAGTATCAcc----------CCTGaaTTGAATaa

71
More methods for multiple alignment

• T-Coffee
• PIMA
• Muscle
• Prrp
• Mafft
• ProbCons

72
Substitution matrices

• Similarity score s(a,b) for amino acids a and b
  based on probability pa,b of substitution a -gt b
• Idea it is more reasonable to align amino acids
  that are often replaced by each other!

73
Substitution matrices

• Assumptions
• pa,b does not depend on sequence position
• Sequence positions independent of each other
• pa,b pb,a (symmetry!)

74
Substitution matrices

• Compute score s(a,b) for degree of similarity
  between amino acids a and b
• Probability pa,b of substitution
• a ? b (or b ? a),
• Frequency qa of a
• Define
• s(a,b) log (pa,b / qa qb)

75

Substitution matrices
76
Substitution matrices

• To calculate pa,b
• Consider alignments of related proteins and count
  substitutions
• a ? b (or b ? a)

77
Substitution matrices

• To calculate pa,b
• Consider alignments of related proteins and count
  substitutions
• a ? b (or b ? a)
• ESWTS-RQWERYTIALMSDQRREVLYWIALY
• ERWTSERQWERYTLALMS-QRREALYWIALY

78
Substitution matrices

• To calculate pa,b
• Consider alignments of related proteins and count
  substitutions
• a ? b (or b ? a)
• ESWTS-RQWERYTIALMSDQRREVLYWIALY
• ERWTSERQWERYTLALMS-QRREALYWIALY

79
Substitution matrices

• Problems involved
• Probability pa,b depends on time t since
  sequences separated in evolution pa,b pa,b
  (t)
• Protein families contain multiple sequences
  phylogenetic tree must be known!
• Alignment of protein families must be known!
• Multiple mutations at one sequence position

80
Substitution matrices

• M. Dayhoff et al., Atlas of Protein sequence and
  Structure, 1978
• PAM matrices

81
Substitution matrices

• Calculation of pa,b(t)
• Consider multiple alignments of closely related
  protein families
• Count occurrence of a and b at corresponding
  positions in alignments using phylogenetic tree
• Estimate pa,b(t) for small times t
• Calculate conditional probabilities p(ab,t) for
  small t
• Normalize to distance 1 PAM ( percentage of
  accepted mutations)
• Calculate p(ab,t) for larger evolutionary
  distances by matrix multiplication
• Calculate pa,b(t) for larger evolutionary
  distances

82
Substitution matrices
83
Substitution matrices

• Alternative BLOSUM matrices
• S. Henikoff and J.G. Henikoff, PNAS, 1992
• Basis BLOCKS database, gap-free regions of
  multiple alignments.
• Cluster of sequences if percentage of similarity
  gt L
• Estimate pa,b(t) directly.
• Default values L 62, L 50