Protein Docking and Interactions Modeling CS 374 Maria Teresa Gil Lucientes November 4, 2004.

Protein Docking and Interactions Modeling

CS 374Maria Teresa Gil Lucientes

November 4, 2004

Overview of the lecture

• Introduction to molecular docking:

• Definition• Types• Some techniques• Programs

• Algorithm for Protein-Protein docking based in paper:

“Protein-Protein Docking with Simultaneous Optimization of Rigid-body Displacement and Side-chain Conformations”

Jeffrey J. Gray, Stewart Moughon, Chu Wang, Ora Schueler-FurmanBrian Kuhlman, Carol A. Rohl and David Baker

J. Mol. Biol. (2003) 331, 281-299

What is Docking?

Docking attempts to find the “best” matching between two molecules

… a more serious definition…

• Given two biological molecules determine:

- Whether the two molecules “interact”

- If so, what is the orientation that maximizes the “interaction” while minimizing the total “energy” of the complex

• Goal: To be able to search a database of molecular structures and retrieve all molecules that can interact with the query structure

Why is docking important?

• It is of extreme relevance in cellular biology, where function is accomplished by proteins interacting with themselves and with other molecular components

• It is the key to rational drug design: The results of docking can be used to find inhibitors for specific target proteins and thus to design new drugs. It is gaining importance as the number of proteins whose structure is known increases

Example: HIV-1 Protease

Active Site(Aspartyl groups)

Example: HIV-1 Protease

Why is this difficult?

• Both molecules are flexible and may alter each other’s structure as they interact:

• Hundreds to thousands of degrees of freedom (DOF)

• Total possible conformations are astronomical

Types of Docking studies

• Protein-Protein Docking• Both molecules usually considered rigid

• 6 degrees of freedom

• First apply steric constraints to limit search space and the examine energetics of possible binding conformations

• Protein-Ligand Docking• Flexible ligand, rigid-receptor

• Search space much larger

• Either reduce flexible ligand to rigid fragments connected by one or several hinges, or search the conformational space using monte-carlo methods or molecular dynamics

Some techniques

• Surface representation, that efficiently represents the docking surface and identifies the regions of interest (cavities and protrusions)

• Connolly surface• Lenhoff technique• Kuntz et al. Clustered-Spheres• Alpha shapes

• Surface matching that matches surfaces to optimize a binding score:

• Geometric Hashing

Surface Representation

• Each atomic sphere is given the van der Waals radius of the atom

• Rolling a Probe Sphere over the Van der Waals Surface leads to the Solvent Reentrant Surface or Connolly surface

Lenhoff technique

• Computes a “complementary” surface for the receptor instead of the Connolly surface, i.e. computes possible positions for the atom centers of the ligand

Atom centers of the ligand

van der Waals surface

Kuntz et al. Clustered-Spheres• Uses clustered-spheres to identify cavities on the receptor and

protrusions on the ligand

• Compute a sphere for every pair of surface points, i and j, with the sphere center on the normal from point i

• Regions where many spheres overlap are either cavities (on the receptor) or protrusions (on the ligand)

i

j

Alpha Shapes

• Formalizes the idea of “shape”

• In 2D an “edge” between two points is “alpha-exposed” if there exists a circle of radius alpha such that the two points lie on the surface of the circle and the circle contains no other points from the point set

Alpha Shapes: Example

Alpha=infinity

Alpha=3.0 Å

Surface Matching

• Find the transformation (rotation + translation) that will maximize the number of matching surface points from the receptor and the ligand

First satisfy steric constraints…• Find the best fit of the receptor and ligand using only geometrical

constraints

… then use energy calculations to refine the docking• Selet the fit that has the minimum energy

Geometric Hashing

Building the Hash Table:– For each triplet of points from the ligand, generate a unique

system of reference

– Store the position and orientation of all remaining points in this coordinate system in the Hash Table

Searching in the Hash Table– For each triplet of points from the receptor, generate a unique

system of reference

– Search the coordinates for each remaining point in the receptor and find the appropriate hash table bin: For every entry there, vote for the basis

Geometric Hashing

– Determine those entries that received more than a threshold of votes, such entry corresponds to a potential match

– For each potential match recover the transformation T that results in the best least-squares match between all corresponding triplets

– Transform the features of the model according to the recovered transformation T and verify it. If the verification fails, choose a different receptor triplet and repeat the searching.

Docking Programs

More information in: http://www.bmm.icnet.uk/~smithgr/soft.html

The programs are:

• DOCK (I. D. Kuntz, UCSF)

• AutoDOCK (Arthur Olson, The Scripps Research Institute)

• RosettaDOCK (Baker, Washington Univ., Gray, Johns Hopkins Univ.)

DOCK

DOCK works in 5 steps:

• Step 1 Start with crystal coordinates of target receptorStep 1 Start with crystal coordinates of target receptor

• Step 2 Generate molecular surface for receptor

• Step 3 Generate spheres to fill the active site of the receptor: The spheres become potential locations for ligand atoms

• Step 4 Matching: Sphere centers are then matched to the ligand atoms, to determine possible orientations for the ligand

• Step 5 Scoring: Find the top scoring orientation

DOCK: Example

1 2

3

- HIV-1 protease is the target receptor- Aspartyl groups are its active side

DOCK

DOCK works in 5 steps:

• Step 1 Start with crystal coordinates of target receptor

• Step 2 Generate molecular surface for receptorStep 2 Generate molecular surface for receptor

• Step 3 Generate spheres to fill the active site of the receptor: The spheres become potential locations for ligand atoms

• Step 4 Matching: Sphere centers are then matched to the ligand atoms, to determine possible orientations for the ligand

• Step 5 Scoring: Find the top scoring orientation

DOCK: Example

1 2

3

- HIV-1 protease is the target receptor- Aspartyl groups are its active side

DOCK

DOCK works in 5 steps:

• Step 1 Start with crystal coordinates of target receptor

• Step 2 Generate molecular surface for receptor

• Step 3 Generate spheres to fill the active site of the Step 3 Generate spheres to fill the active site of the receptor: The spheres become potential locations for receptor: The spheres become potential locations for ligand atomsligand atoms

• Step 4 Matching: Sphere centers are then matched to the ligand atoms, to determine possible orientations for the ligand

• Step 5 Scoring: Find the top scoring orientation

DOCK: Example

1 2

3

- HIV-1 protease is the target receptor- Aspartyl groups are its active side

DOCK

DOCK works in 5 steps:

• Step 1 Start with crystal coordinates of target receptor

• Step 2 Generate molecular surface for receptor

• Step 3 Generate spheres to fill the active site of the receptor: The spheres become potential locations for ligand atoms

• Step 4 Matching: Sphere centers are then matched to Step 4 Matching: Sphere centers are then matched to the ligand atoms, to determine possible orientations for the ligand atoms, to determine possible orientations for the ligandthe ligand

• Step 5 Scoring: Find the top scoring orientation

DOCK: Example

4 5

• Three scoring schemes: Shape scoring, Electrostatic scoring and Force-field scoring• Image 5 is a comparison of the top scoring orientation of the molecule thioketal with the orientation found in the crystal structure

DOCK

DOCK works in 5 steps:

• Step 1 Start with crystal coordinates of target receptor

• Step 2 Generate molecular surface for receptor

• Step 3 Generate spheres to fill the active site of the receptor: The spheres become potential locations for ligand atoms

• Step 4 Matching: Sphere centers are then matched to the ligand atoms, to determine possible orientations for the ligand

• Step 5 Scoring: Find the top scoring orientationStep 5 Scoring: Find the top scoring orientation

DOCK: Example

4 5

• Three scoring schemes: Shape scoring, Electrostatic scoring and Force-field scoring• Image 5 is a comparison of the top scoring orientation of the molecule thioketal with the orientation found in the crystal structure

Other Docking programs

AutoDock– AutoDock was designed to dock flexible ligands into receptor

binding sites

– The strongest feature of AutoDock is the range of powerful optimization algorithms available

RosettaDOCK– It models physical forces and creates a very large number of

decoys

– It uses degeneracy after clustering as a final criterion in decoy selection

CAPRI Challenge (2002)

• At least one docking partner presented in its unbound form

• Participants permitted 5 attempts for each target

The 7 CAPRI Docking Targets

CAPRI ChallengeParticipants & Algorithms

Results: CAPRI Challenge

This were the results for the different predictors and targets:

A Protein-Protein Docking Algorithm (Gray & Baker)

• Our goal is to try to predict protein-protein complexes from the coordinates of the unbound monomer components.

• The method is divided in two steps: A low-resolution Monte Carlo search and a final optimization using Monte Carlo minimization.

• Up to 105 independent simulations are carried out, and the resulting “decoys” are ranked using an energy function.

• The top-ranking decoys are clustered to select the final predictions.

Docking protocol

Docking protocol: Step 1

RANDOM START POSITIONRANDOM START POSITION

• Creation of a decoy begins with a random orientation of each partner and a translation of one partner along the line of protein centers to create a glancing contact between the proteins

Docking protocol

Docking protocol: Step 2

LOW-RESOLUTION MONTE CARLO SEARCHLOW-RESOLUTION MONTE CARLO SEARCH

• One partner is translated and rotated around the surface of the other through 500 Monte Carlo move attempts

• We use a low-resolution representation: N, C, C, O for the backbone and a “centroid” for the side-chain

• The score is based in the correctness of each decoy: A reward contacting residues, a penalty for overlapping residues, an alignment score, residue environment and residue-residue interacions terms

Docking protocol

Docking protocol: Step 3

HIGH-RESOLUTION REFINEMENTHIGH-RESOLUTION REFINEMENT

• Explicit side-chains are added to the protein backbones using a rotameter packing algorithm, thus changing the energy surface

• An explicit minimization finds the nearest local minimum accessible via rigid body translation and rotation

• Start and Finish positions are compared by the Metropolis criterion

Docking protocol

Docking protocol: Step 3

• Before each cycle, the position of one protein is perturbed by random translations and by random rotations

• To simultaneously optimize the side-chain conformations and the rigid body position, the side-chain packing and the minimization operations are repeated 50 times

Docking protocol: Step 3

COMPUTATIONAL EFFICIENCY

1. The packing algorithm usually varies the conformation of only one residue at a time: A combinatorial rotamer optimization is performed only once every eight cycles

2. A filter is employed periodically to detect inferior decoys and and reject them without further refinement

Docking protocol

Docking protocol: Step 4

CLUSTERING & PREDICTIONSCLUSTERING & PREDICTIONS

• The search procedure is repeated to create approximately 105 decoys per target

• The 200 best-scoring decoys are then clustered on the basis of the root-mean-squared distance (rmsd) using a hierarchical clustering algorithm

• The clusters with the most members are selected as the final predictions and ranked according to cluster sizes

Docking protocol: Results

Conclusions

• The so-called computational molecular docking problem is far from being solved. There are two major bottle-necks:

1. The algorithms can handle only a limited extent of backbone flexibility

2. The availability of selective and efficient scoring functions

… and Thanks!! Questions??