What you should know by now

Concepts:

• Pairwise alignment

• Global, semi-global and local alignment

• Dynamic programming

• Sequence similarity (Sum-of-Pairs)

• Profiles (a basic understanding)

Biological Definitions for Related Sequences

• Homologues are similar sequences in two different organisms that have been derived from a common ancestor sequence. Homologues can be described as either orthologues or paralogues.

• Orthologues are similar sequences in two different organisms that have arisen due to a speciation event. Orthologs typically retain identical or similar functionality throughout evolution.

• Paralogues are similar sequences within a single organism that have arisen due to a gene duplication event.

• Xenologues are similar sequences that do not share the same evolutionary origin, but rather have arisen out of horizontal transfer events through symbiosis, viruses, etc.

So this means …

 

Source: http://www.ncbi.nlm.nih.gov/Education/BLASTinfo/Orthology.html

Multiple Sequence Alignment

Sequences can be mutated or rearranged to perform an altered Sequences can be mutated or rearranged to perform an altered function. function.

which changes in the sequence have caused a change in the which changes in the sequence have caused a change in the functionality.functionality.

Multiple sequence alignment: the idea is to take three or Multiple sequence alignment: the idea is to take three or more sequences and align them so that the greatest more sequences and align them so that the greatest number of similar characters are aligned in the same number of similar characters are aligned in the same column of the alignment.column of the alignment.

•hold information about which regions have high mutation rates hold information about which regions have high mutation rates over evolutionary time and which are evolutionarily conserved over evolutionary time and which are evolutionarily conserved •identification of regions or domains that are critical to functionality.identification of regions or domains that are critical to functionality.

Sequences can be conserved across species and perform similar or Sequences can be conserved across species and perform similar or identical functions.identical functions.

What to ask yourself

• How do we get a multiple alignment? (three or more sequences)

• Which way is best?– Do we go for max accuracy, least

computational time or the best compromise?

• What do we want to achieve each time?

Multiple alignment profilesGribskov et al. 1987

ACDWY

-

i

fA..fC..fD..fW..fY..Gapo, gapxGapo, gapx

Position dependent gap penalties

Core region Core regionGapped region

Gapo, gapx

fA..fC..fD..fW..fY..

fA..fC..fD..fW..fY..

Profile building

ACDWY

Gappenalties

i0.30.100.30.3

0.51.0

Position dependent gap penalties

0.50000.5

00.50.20.10.2

1.0

Example: Each aa is represented as a frequency, penalties as weights

ACD……VWY

sequence

profile

Profile-sequence alignment

ACD..Y

ACD……VWY

profile

profile

Profile-profile alignment

Multiple alignment methods

• Multi-dimensional dynamic programming• Progressive alignment• Iterative alignment

Simultaneous multiple alignmentMulti-dimensional dynamic programming

The combinatorial explosion:• 2 sequences of length n

– n2 comparisons

• Comparison number increases exponentially– i.e. nN where n is the length of the sequences, and N is the

number of sequences–

• Impractical for even a small number of short sequences quite quickly

Multi-dimensional dynamic programming (Murata et al, 1985)

Sequence 1

Seq

uenc

e 2

Sequence 3

The MSA approach

MSA (Lipman et al., 1989, PNAS 86, 4412)• Calculate all pairwise alignment scores (SP). • Use the scores to predict a tree. • Calculate pair weights based on the tree. • Produce a heuristic alignment based on the tree. • Calculate the maximum weight for each sequence pair. • Determine the spatial positions that must be calculated to obtain

the optimal alignment. • Perform the optimal alignment. • Report the weight found compared to the maximum weight

previously found• extremely slow and memory intensive• Max 8-9 sequences of ~250 residues

The DCA approach

DCA (Stoye et al 1997) • Iteratively split at

optimal cut points• Use MSA • Concatenate

So in effect …Sequence 1

Seq

uenc

e 2

Sequence 3

Multiple alignment methods

• Multi-dimensional dynamic programming

• Progressive alignment

• Iterative alignment

Progressive alignment

1) Perform pairwise alignments of all of the sequences2) Use the alignment scores to produces a dendrogram using

neighbour-joining methods3) Align the sequences sequentially, guided by the

relationships indicated by the tree

• Biopat (first method ever)• MULTAL (Taylor 1987)• DIALIGN (1&2, Morgenstern 1996)• PRRP (Gotoh 1996)• Clustal (Thompson et al 1994)• Praline (Heringa 1999)• T Coffee (Notredame 2000)• POA (Lee 2002)

Progressive multiple alignment1213

45

Guide tree Multiple alignment

Score 1-2

Score 1-3

Score 4-5

Scores Similaritymatrix5×5

Scores to distances Iteration possibilities

General progressive multiple alignment technique (follow generated tree)

13

25

13

13

13

25

254

d

root

Praline progressive strategy

13

2

13

13

13

25

254

d

4

There are problems:

Accuracy is very important !!!! Errors are propagated into the progressive

steps

“ Once a gap, always a gap”Feng & Doolittle, 1987

Pair-wise alignment quality versus sequence identity

(Vogt et al., JMB 249, 816-831,1995)