Structural alphabets, from protein structure description - PowerPoint PPT Presentation

1 / 28
About This Presentation
Title:

Structural alphabets, from protein structure description

Description:

SCOP 95. database. Top Hits. Query using DP. Score based ranking ... Mining of structurally similar proteins from SCOP at 95% database ... – PowerPoint PPT presentation

Number of Views:122
Avg rating:3.0/5.0
Slides: 29
Provided by: dsimbI
Category:

less

Transcript and Presenter's Notes

Title: Structural alphabets, from protein structure description


1
Structural alphabets, from protein structure
description to comparison
Manoj Tyagi LBGM Université de la Réunion
2
Synopsis
  • Classical method of protein structure analysis
  • Structural alphabets protein blocks
  • Reduction of protein space
  • Generation of PB substitution matrix
  • Applications of PB matrix
  • Structure comparison and mining
  • Identification of rigid body movement or local
    conformation changes
  • Conclusion

3
Analysis of protein structures
4
Secondary structures 3 states
5
Structural Alphabets / Protein Blocks
  • A structural alphabet is a set (or library) of
    small structural motifs/prototypes which
    approximate every part of the protein structures.
  • They are composed by a limited number of
    recurrent structural elements of proteins. i.e.
    are recurring in protein space.
  • Protein blocks are set of 16 short structural
    motifs of 5 consecutive residues/amino acids.
  • Each motif is represented by vector/set of 8
    dihedral angles (f,y).
  • Denoted by letter a, b, ,p.

6
Structural Alphabets / Protein Blocks
  • PBs can be characterized by their secondary
    structure composition. e.g.PB m forms the central
    part of helix and PB d is ideal for sheet
  • PBs from a to c and d to f are mainly concerned
    with N and C caps of sheet
  • Similarly PBs k, l and n, o, p form N and C
    caps of helix.
  • PBs labelled from g to j are mainly concerned
    with coils.

7
yn-2 , fn-1 (41.14,75.53) yn-1 , fn
(13.92,99.80) yn , fn1 (131.88,96.27) yn1 ,
fn2 (122.08,99.68)
8
Encoding 3D structure as 1D PB sequence
PB sequence
KBCCDDDDFBFKLMMMMMMMMMMNOPABDCDDFBFKL MMMMMMNGOIAB
DCDDFBDGHILMLMMMMMMMMPMKL MMPCCDDDDFBDCFKLMMMMMMNO
PABDCDDDDDFKL MMMMMMMMMMMNO
What we found ? PBs when combined together gives
back regular structures of a protein and also
highlight variable regions present between
regular structure elements
3D structure
9
What can we do with PB sequence ?
  • PB sequence comparison to identify equivalent
    regions
  • Structure comparison based on sequence alignment
    algorithm
  • Identification of conformational change or rigid
    body displacement/shift in proteins
  • Extension of above to study active inactive
    states of enzymes

What do we need now? A substitution matrix !
10
Generation of PBs substitution matrix
  • Required large number of structurally aligned
    proteins
  • PALI database was used
  • Structure alignments encoded into PB sequence
    alignments
  • Calculations of substitution frequency of each
    PB in conservered regions
  • Conversation to log odds score

Balaji S, Sujatha S, Kumar SS, Srinivasan N.
PALI-a database of Phylogeny and ALIgnment of
homologous protein structures. Nucleic Acids Res
200129(1)61-65.
11
Substitution table calculation
PALI aa alignment
PB alignment
Substitution count
Log odd score
PB matrix (16x16)
12
PB substitution matrix
Tyagi M, Venkataraman SG, Srinivasan N, de
Brevern AG, Offmann B A substitution matrix for
structural alphabet based on structural alignment
of homologous proteins and its applications.
Proteins in press.
13
Potential applications of PB substitution table
  • Use of classical sequence alignment methods (DP)
    to align and compare two PBs sequences to
    identify structurally equivalent and non
    equivalent regions
  • Extension of above to structure mining in large
    database
  • Combination of substitution table with rigid body
    superimposition to identify conformational
    changes or rigid body displacment in homologous
    proteins
  • Extension of above approach to study enzymes
    active/inactive form

14
Structure comparison
3chy
2fox
?
15
Structure comparison
3chy_ KBCCDDDDFBF--KLMMMMMMMMMMNOPABDCDDFBFKLMMMM
MMNGOIABDCDDFBDGHI 2fox_ ---DDDDFKOMMMMMMMMMMMMMM
MNOPACDDDD--FKLPCFKLMMM-PFBDCDDDDDEHJ 3chy_
LM--LMMMMMMMMPMKLMMPCCDDDDFBDCFKL-----MM-MMM-M-NOP
A--BDCDDDD- 2fox_ LPACFKLMMMMMMMMMMMMGHIACDDEHIAF
KLNOMMMMMMMMMMMMNOPACFBACDDDEH 3chy_
-DFKLMMMMMMMMMMMNO 2fox_ JACKLMMMMMMMMMMMMM
16
Contd
  • Metallohydrolase superfamily members 1qh5a and
    1smla of equivalent length

17
Contd
  • 1bnka 1fmtb from all beta class FMT
    C-terminal domain like superfamily

18
Structure mining
Score based ranking
  • 98.2 89 of success rate in finding true
    class and fold resp. within top 10 hits

19
Structure mining
Based on 7259 x 7259 pairwise PB alignments
20
Protein Block Expert (PBE)
  • Pair wise PB sequence alignment to compare two
    protein structure
  • Mining of structurally similar proteins from SCOP
    at 95 database
  • Preprocessed pair wise PB alignments at family
    and superfamily level
  • Provides both local and global alignment
    algorithm falvors
  • Available at http//bioinformatics.univ-reunion.fr
    /PBE/

Tyagi M, Sharma P, Swamy CS, Cadet F, Srinivasan
N, Brevern AG, Offmann B Protein Block Expert
(PBE) A web-based protein structure analysis
server using a structural alphabet. Nucl Acids
Res in press.
21
Identification of conformational changes
  • Rigid body superimposition methods e.g. STAMP
    report sequence alignment based on structurally
    equivalent and variable regions
  • A cutoff/threshold residue residue rmsd value is
    used to define variable regions
  • Are these high rmsd values, due to difference in
    conformations or rigid body displacment of
    equivalent regions ?
  • We dont know from simple rigid body structure
    alignment

22
Rigid body superimposition
Corresponding sequence alignment
Two proteins
Study substitution scores in variable regions
PB alignment
-ve
ve
Rigid body displacment
Conformation change
23
  • Analysis of two distantly related tRNA synthetase
    1eiy and 1set

24
  • Example of two zinc metallo proteinases 1hov and
    1fbl

25
Cyclic AMP dependent protein kinase
26
(No Transcript)
27
Conclusion
  • Use of simple PB sequence representation of 3D
    structure enables us to align and compare protein
    structures in simple and intuitive way
  • PB alignment enables us to identify structurally
    equivalent regions and highlights subtle
    differences
  • Efficient and fast mining process gt structural
    genomics
  • Methodology is simple and can be used in other
    applications e.g. realignment of variable regions
    from rigid body superimposition alignment

28
Thank you
Write a Comment
User Comments (0)
About PowerShow.com