Structure and Function Suggested readings: Lee D, Redfern O, Orengo C. Predicting protein function from sequence and structure. Nat Rev Mol Cell Biol.

Structure and Function

Suggested readings:

Lee D, Redfern O, Orengo C.Predicting protein function from sequence and structure. Nat Rev Mol Cell Biol. 2007 8(12):995-1005.

Chapter 2 of the book Understanding bioinformatics Marketa Zvelebil Jeremy O. Baum

Wu CH, Apweiler R, Bairoch A, Natale DA, Barker WC, Boeckmann B, Ferro S,Gasteiger E, Huang H, Lopez R, Magrane M, Martin MJ, Mazumder R, O'Donovan C,Redaschi N, Suzek B.The Universal Protein Resource (UniProt): an expanding universe of proteininformation. Nucleic Acids Res. 2006 Jan 1;34(Database issue):D187-91.

Kinoshita K, Nakamura H. Protein informatics towards function identification.Curr Opin Struct Biol. 2003 Jun;13(3):396-400. Review.

The information on biological data is not stored in books…..

BUT

in Electronic Databases publicly available (generally!) and accessibleby internet

Most of the work of a bioinformatician is to learn how to retrieve those data,

how to analyse them, and how to produce novel data necessary to the understanding of the complex biological world

“Empirical art”

In 1965 Gordon Moore, co-founder of Intel noticed that:Every chip had a capacity double to the predecessor and that every 18-24 months a new generation of chip was born.

In the seventies Dickerson, a professor of Physical Chemistrynoticed that the number of solved X-ray structures of proteins

had increased from 1 in 1961 to 23 in 1977

After the completion of the human genome project we have availablemillions of sequences which represent a biological ‘knowledge’ thatwe have to understand…

Growth of pdb structuresStructures are collected in the PDB databank:http://www.rcsb.org/pdb/home/home.do

So our ‘knowledge’ is very limited

45.000

Protein Structure Prediction

• In theory, a protein structure can be solved computationally

• A protein folds into a 3D structure to minimizes its free potential energy

• The problem can be formulated as a search problem

for minimum energy• the search space is enormous• the number of local minima increases exponentially

Computationally it is an exceedingly difficult problem

8

protein prediction .vs. protein determination

Exp

erim

enta

l da

ta

inferred data

X-Ray

NMR

Comparative Modeling

Threading

Ab-initio

QuickTime™ and a decompressor

are needed to see this picture.

….from a string of pearl to a folded structure….

11

Why is it useful to know the structure of a protein, not only its sequence?

• The biochemical function (activity) of a protein is defined by its interactions with other molecules.

• The biological function is in large part a consequence of these interactions.

• The 3D structure is more informative than sequence because interactions are determined by residues that are close in space but are frequently distant in sequence.

In addition, since evolution tends to conserve function and function depends more directly on structure than on sequence, structure is more

conserved in evolution than sequence.

The net result is that patterns in space are frequently more recognizable than patterns

in sequence.

• Seq. Id. > 50%: core region ~90% of the structure, r.m.s.d. of themain chain around 1.0 Å• Seq. Id. < 20%: core region ~50% of thestructure, r.m.s.d. of main chain around 1.8 Å

[Chothia & Lesk, EMBO J.(1986) 5: 823-826]

There is a relationship between sequence similarity and structural similarity

A Ramachandran plot (also known as a Ramachandran map or a Ramachandran diagram), is a way to visualize dihedral angles φ against ψ of amino acid residues in protein structure. It shows the possible conformations of φ and ψ angles for a polypeptide.Mathematically, the Ramachandran plot is the visualization of a function (torus). Hence, the conventional Ramachandran plot is a projection of the torus on the plane, resulting in a distorted view and the presence of discontinuities.One would expect that larger side chains would result in more restrictions and consequently a smaller allowable region in the Ramachandran plot. In practice this does not appear to be the case; only the methylene group at the β position has an influence. Glycine has a hydrogen atom, with a smaller van der Waals radius, instead of a methyl group at the β position. Hence it is least restricted and this is apparent in the Ramachandran plot for Glycine for which the allowable area is considerably larger.In contrast, the Ramachandran plot for proline shows only a very limited number of possible combinations of ψ and φ.

QuickTime™ and a decompressor

are needed to see this picture.

A Ramachandran plot generated from the protein PCNA, a human DNA clamp protein that is composed of both beta sheets and alpha helices (PDB ID 1AXC). Points that lie on the axes indicate N- and C-terminal residues for each subunit. The green regions show possible angle formations that include Glycine, while the blue areas are for formations that don't include Glycine.

From Wikipedia

The DSSP code

H = alpha helix B = residue in isolated beta-bridge E = extended strand, participates in beta ladder G = 3-helix (3/10 helix) I = 5 helix (pi helix) T = hydrogen bonded turn S = bend

Dictionary of protein secondary structure: pattern recognition of hydrogen-bonded and geometrical features.

Biopolymers. 1983 Dec;22(12):2577-637.

http://swift.cmbi.ru.nl/gv/dssp/

http://bioweb.pasteur.fr/seqanal/interfaces/dssp-simple.html

The secondary structure assignment with DSSP over a database of structures can be used as ‘standard of truth’ for secondary structure prediction methods.

the alpha helix (α-helix) in which every backbone N-H group donates a hydrogen bond to the backbone C=O group of the amino acid four residues earlier (             hydrogen bonding).

The amino acids in a 310-helix are arranged in a right-handed helical structure. The N-H group of an amino acid forms a hydrogen bond with the C = O group of the amino acid three residues earlier;this repeated i + 3 → i hydrogen bonding

defines a 310-helix.

Homology: two proteins are homologous

if they have a common ancestor. This is a binary property (once defined is either there or not).

It is a useful information:when a known gene G is homologous to an unknown gene X we will gain information on X from G through transitivity.

OrthologyHomologous sequences are orthologous if they were separated by a speciation event: when a species diverges into two separate species, the divergent copies of a single gene in the resulting species are said to be orthologous. Orthologs, or orthologous genes, are genes in different species, that are similar to each other because they originated from a common ancestor.

ParalogyHomologous sequences are paralogous if they were separated by a gene duplication event: if a gene in an organism is duplicated to occupy two different positions in the same genome, then the two copies are paralogous.A set of sequences that are paralogous are called paralogs of each other. Paralogs typically have the same or similar function, but sometimes do not: due to lack of the original selective pressure upon one of copy of the duplicated gene, this copy is free to mutate and acquire new functions.

Homologies:Rouse

Hemoglobin

Ratα Hb

Mouseα Hb

Mouseβ Hb

Ratβ Hb

OrthologsParalogs

α Hemoglobin β Hemoglobin

Geneduplication

Rouse RouseSpeciation

The genes encoding myoglobin and hemoglobin are considered to be ancient paralogs. Similarly, the four known classes of hemoglobins (hemoglobin A, hemoglobin A2, hemoglobin S, and hemoglobin F) are paralogs of each other.

While each of these genes serve the same basic function of oxygen transport, they have already diverged slightly in function: fetal hemoglobin (hemoglobin F) has a higher affinity to oxygen than adult hemoglobin.Paralogous genes often belong to the same species, but this is not necessary: for example, the hemoglobin gene of humans and the myoglobin gene of chimpanzees are paralogs. This is a common problem in bioinformatics: when genomes of different species have been sequenced and homologous genes have been found, one can not immediately conclude that these genes have the same or similar function, as they could be paralogs whose function has diverged.

QuickTime™ and a decompressor

are needed to see this picture.

Where it all starts…..http://www.expasy.ch/sprot/

sequence

structure

function

-

+ con

serv

atio

n

Function is attributed to very few atoms absolutely

conserved during the evolutionary process

The ‘centrality’ of a 3D structure

How Can We Compare Sequences ?The Twilight Zone

Length

%Sequence Identity

100

Same 3D Fold

Twilight Zone

Similar SequenceSimilar Structure

30%

Different SequenceStructure ????

30

How Do Sequences Evolve ?

In a structure, each Amino Acid plays a Special Role

In the core, SIZE MATTERS

On the surface, CHARGE MATTERS

--+

Divergent evolution?

Convergent evolution?

Inactivation of neurotransmitters

Sensorial cellular response

sequence identityfor 104 over 198 residues

Structures are collected in the PDB databank http://www.rcsb.org/pdb/home/home.do

The Protein Data Bank (PDB) is a repository for 3-D structural data of proteins and nucleic acids. This data, typically obtained by X-ray crystallography or NMR spectroscopy, is submitted by biologists and biochemists from around the world, is released into the public domain, and can be accessed for free. The database is the central repository for biological structural data.

PDB COORDINATE FILE FORMAT

Figure 1: Each separate column in a given section of a PDB file is designated as a different "field”

#Definition of PDB format:#http://www.wwpdb.org/documentation/format23/v2.3.html#The ATOM record:##COLUMNS DATA TYPE FIELD DEFINITION#---------------------------------------------------------------------------------# 1 - 6 Record name "ATOM "# 7 - 11 Integer serial Atom serial number.#13 - 16 Atom name Atom name.#17 Character altLoc Alternate location indicator.#18 - 20 Residue name resName Residue name.#22 Character chainID Chain identifier.#23 - 26 Integer resSeq Residue sequence number.#27 AChar iCode Code for insertion of residues.#31 - 38 Real(8.3) x Orthogonal coordinates for X# in Angstroms.#39 - 46 Real(8.3) y Orthogonal coordinates for Y# in Angstroms.#47 - 54 Real(8.3) z Orthogonal coordinates for Z# in Angstroms.#55 - 60 Real(6.2) occupancy Occupancy.#61 - 66 Real(6.2) tempFactor Temperature factor.#73 - 76 LString(4) segID Segment identifier,# left-justified.#77 - 78 LString(2) element Element symbol,# right-justified.#79 - 80 LString(2) charge Charge on the atom.

Atomic coordinates

Contact plots

Calculate the distances between C-beta atoms

Define a distance cut-off (7 Angstrom)

If distance is shorter than cut-off define it as a contact

$atomDist = sqrt(($xCoor[$cBeta[$i]] - $xCoor[$cBeta[$j]])**2 + ($yCoor[$cBeta[$i]] - $yCoor[$cBeta[$j]])**2 + ($zCoor[$cBeta[$i]] - $zCoor[$cBeta[$j]])**2);

Human prion protein Ovine prion protein variant R168

Ovine prion Sheep prion

Divergent evolution?

Divergent evolution?

Convergent evolution?

Inactivation of neurotransmitters

Sensorial cellular response

Convergent evolution?

QuickTime™ and a decompressor

are needed to see this picture.

Proteins are organised in domains with specific architecture

So proteins with similar function have similar domain architecture!!

Domains: conserved folded units stable in isolation

DTGM Fusion Protein to Treat Patients With Recurrent or Refractory Acute Myeloid LeukemiaAnticancer drug formed by the combination of diphtheria toxin and a colony- stimulating factor (GM-CSF)

Calmodulin

http://structbio.vanderbilt.edu/cabp_database/pic_gallery/ca_coord/cam_chapter1.html

EF-hand Sequencesin Pfam Database

-Z-X#ZYX1 3 5 7 9 12

Protein function can be divided into three broad areas: molecular function, biological process and cellular component.

Molecular function describes activity at the molecular level, such as catalysis, which is commonly predicted through methods that identify homologues or orthologues.

Biological process describes broader functions that are carried out by assemblies of molecular functions, such as a particular metabolic pathway.

Genomic inference methods can identify the direct physical protein–protein interactions and indirect functional associations that are found in biological processes.

Cellular component describes the compartment(s) of a cell in which the protein performs its function. This component can be predicted through methods that predict signal sequences, residue composition, membrane association or post-translational modifications.

Within these areas there are broad categories of computational methods, all of which ultimately depend on experimental data.

From Lee D, Redfern O, Orengo C.

Predicting protein function from sequence and structure. Nat Rev Mol Cell Biol. 2007 8(12):995-1005

What does it means function?EC nomenclature

http://www.geneontology.org/GO.annotation.shtml

http://www.blast2go.de/

Go project provides three structured vocabularies (ontologies) to describe gene products in terms of their 1)associated biological processes; 2)cellular components; 3)molecular functions

From Lee D, Redfern O, Orengo C.

Predicting protein function from sequence and structure. Nat Rev Mol Cell Biol. 2007 8(12):995-1005

Learning outcomes of the lecture

The relevance of the knowledge of a 3D structure: why? Implications?

What are the most common secondary structure elements

What are the ‘teachings’ of the PDB database

How can we approach the search of the function of a protein or a gene