1

Re-usable components forRe-usable components forstructural structural bioinformaticsbioinformatics

Sophie COON

The Scripps Research InstituteLa Jolla, California

SCHEDULESCHEDULE

o I – Code development strategieso II – Re-usable components

? MolKit, DejaVu, ViewerFramework

o III - From building blocks to applications? PMV

o Conclusion

2

Protein-Protein-LigandLigand

Complex AssembliesComplex AssembliesProtein-ProteinProtein-Protein

Molecular SurfacesMolecular SurfacesAutoDockAutoDock

SurfDockSurfDock

SymGenSymGen

MSMSMSMS

HARMONYHARMONY

SymSearchSymSearch

The challengeThe challenge

VisualizationVisualization

DockingDockingMethodsMethods

FoldingFolding

ProteinProteinEngineeringEngineering

SequenceSequenceAnalysisAnalysis

ModelingModeling

MM - MDMM - MD

Ab InitioAb InitioMethodsMethods

ElectrostaticsElectrostaticsCalculationsCalculations

MolecularMolecularSurfacesSurfaces

Etc ...Etc ...

3

“Traditional” solution“Traditional” solution

VisualizationVisualization

DockingDockingMethodsMethods

FoldingFolding

ProteinProteinEngineeringEngineering

SequenceSequenceAnalysisAnalysis

ModelingModeling

MM - MDMM - MD

Ab InitioAb InitioMethodsMethods

ElectrostaticsElectrostaticsCalculationsCalculations

MolecularMolecularSurfacesSurfaces

Etc ...Etc ...

“Application” centric approach“Application” centric approach

Parsing a moleculefile

Dockingmethods

Molecularsurface

ElectrostaticsCalculations

visualization

Ab Initio Methods

4

Our solutionOur solutionHigh level language as a scripting environmentHigh level language as a scripting environment

MolecularSurfaces

Molecular

MolecularSurfaces

Surfaces

Molecules

Molecules

Molecules

DataBase

DataDataBaseBaseElectrostatics

Electrostatics

Electrostatics

Delaunay

DelaunayDelaunay

Homology

Homology

HomologyCSGCSGCSG

3DViewer

3D3DViewerViewer

MM-MDMM-MDMM-MD

NewMethod

NewNewMethodMethod

Your

MethodYourYour

MethodMethod

Writing an applicationWriting an application

Molecules

Molecules

Molecules3DViewer

3D3DViewerViewer

MolecularSurfaces

Molecular

MolecularSurfaces

Surfaces

Electrostatics

Electrostatics

Electrostatics

oo High level codingHigh level codingoo Code re-useCode re-useoo modularitymodularityoo ExtensibleExtensible

5

II - Re-usable componentsII - Re-usable components

o MolKito DejaVuo ViewerFramework

MolKitMolKit

NumericNumeric

MolKitMolKitTreeNode

.parent

.children

...

TreeNode

adopt(child)

TreeNodeSet(ListSet)

[TreeNode, TreeNode, … ].__getattr__( )

.top

.elementType

TreeNode

[ TreeNode1.attr, TreeNode2.attr, …]returns

.name

6

TreeNodeTreeNode, , TreeNodeSetTreeNodeSet

o TreeNodeSet:? Boolean operation? uniq( )? split( )? sort( )? NodesFromName( )? findChildrenOfType( )? findParentOfType( )? …

o TreeNode:? adopt( ) / remove( )? full_name( )? NodeFromName( )? split( ) / merge( )? getParentOfType( )? findType( )? compare( )? assignUniqIndex( )? isAbove( ) / isBelow( )? …

TreeNode TreeNode and and TreeNodeSet TreeNodeSet specializationspecialization

Residue Chain Protein ...

Molecule Atom

Helix Strand Turn Coil

SecondayStructure ...

TreeNode

ResidueSet ChainSet ProteinSet ...

MoleculeSet AtomSet

HelixSet StrandSet TurnSet CoilSet

SecondayStructureSet ...

TreeNodeSet

7

ParsersParsers

PDBMol2PQR...

Parser

MoleculeSet from MolKit.pdbParser import PdbParserparser = PdbParser(‘1crn.pdb’)mols = parser.parse( )

PDB parser MOL2 parserMolecule

Chain

Residue

Atom

Molecule

Chain

Residue

Atom

Molecule

Residue

Atom

Molecule

Atom

ExamplesExamples>>> from MolKit import Read>>> molecules = Read(‘./1crn.pdb’)>>> mol = molecules[0] # Read returns a ProteinSet

>>> print mol.chains.residues.name>>> print mol.chains.residues.atoms[20:85].full_name()

>>> from MolKit.molecule import Atom>>> allAtoms = mol. findType(Atom)>>> set1 = allAtoms.get(lambda x: x.temperatureFactor >20)

>>> allResidues = allAtoms.parent.uniq()>>> import Numeric>>> for r in allResidues:. . . coords = r.atoms.coords. . . r.geomCenter = Numeric.sum(coords) / len(coords)

8

DejaVuDejaVu

TkinterTkinter

NumericNumericPyOpenGLPyOpenGL

DejaVuDejaVu

from DejaVu import Viewervi = Viewer( )from DejaVu.Spheres import Spherescenters = =[[0,0,0],[3,0,0],[0,3,0]] s = Spheres(‘sph’ , centers = centers)s.Set(quality=10)vi.AddObject(s)

FeaturesFeatures

o OpenGL Lighting and Material modelo Object hierarchy with transformation and

rendering properties inheritanceo Arbitrary clipping planeso Material editoro DepthCueing (fog), global anti-aliasingo glSscissors/magic lenso Multi-level pickingo Extensible set of geometries

9

GeometriesGeometries

Geom

o PolyLineo Pointso Sphereso Labelso Arc3D...

IndexedGeoms

o IndexedPolyLineso IndexedPolygomso Triangle_Stripo Quad_Stripo Cylinders ….

DejaVu DejaVu and and MolKitMolKit

>>> from MolKit import Read>>> molecules = Read(‘./1crn.pdb’)>>> mol = molecules[0] # Read returns a ProteinSet>>> coords = mol.chains.residues.atoms.coords>>> radii = mol.defaultRadii()>>> from DejaVu.Spheres import Spheres>>> from DejaVu import Viewer>>> vi = Viewer()>>> sph = Spheres(‘sph’, centers = coords, radii = radii, quality=10)>>> vi.AddObject(sph)

10

ViewerFrameworkViewerFramework

NumericNumericPyOpenGLPyOpenGL TkinterTkinter

DejaVuDejaVu

IdleIdle

ViewerFrameworkViewerFramework

Design featuresDesign features

o Dynamic loading of commandso Python shell for scriptingo Dual interaction mode (GUI/Shell)o Support for command:

? development, logging, GUI, dependencieso Lightweight commands: Macroso Dynamic commands (introspection)o Extensible set of commands

11

ImplementationImplementation

GeomContainer

VFGUI

VFCommandsGUIGUIcommandsobjectsuserpref

UserPreference

addObject( … )removeObject( … ) addCommand( … )askFileOpen( … )askFileSave( … )getUserInput( … )

VFCommand

Objects

ViewerFramework

III - III - From Building Blocks to applicationsFrom Building Blocks to applications

o PMV: Python molecular viewer

12

PMVPMV

NumericNumeric

PyOpenGLPyOpenGL TkinterTkinter

DejaVuDejaVuIdleIdle

ViewerFrameworkViewerFramework

PynchePynche

PmvPmv

MolKitMolKit

Specializing Specializing ViewerFrameworkViewerFrameworkViewerFramework

geomContainer

molGeomContainer

PMV

VFCommand . vf

MoleculeViewer .Molecules

ViewerFramework .Objects

MVCommand .getNodesByMolecules(nodes) .vf

13

Specializing Specializing ViewerFramework ViewerFramework (cont'd)(cont'd)

VFCommandShellCommand

loadCommandCommand

ExitCommand

helpCommandCommand

Etc...

MVCommand

PDBReader

MoleculeLoader

Mol2Reader

DisplayLines

Etc...

ViewerFrameworkPMV

PMVPMV

NumericNumeric

PyOpenGLPyOpenGL TkinterTkinter

DejaVuDejaVuIdleIdle

ViewerFrameworkViewerFramework

PynchePynche

PmvPmv

MolKitMolKit

DEMODEMO

14

ConclusionConclusion

o Validity of the approacho Availabilityo Future directions

Validity of the approachValidity of the approach

o Set of components? extensible? inter-operable? re-usable? short development cycle

o User base expanding beyond our lab.o Components re-use outside the field

of structural biology

15

AvailabilityAvailability

o Modularity enables fine grainlicensing schemes (a la carte)

o Core modules are freely available

o Online Download site:http://www.scripps.edu/~sanner/Python

Future directionsFuture directions

o Add support for editing molecularstructures (i.e. mutations, deletion,addition)

o Interface with MMTK, Tinker, ….o Enhance documentation and tutorialso Setup a CVS server for programmers

wanting to help !o Too many to list ….

16

AcknowledgmentsAcknowledgments

o Michel Sannero Christian Carrillo, Kevin Chano Ruth Huey, Fariba Fanao Vincenzo Tchinke, Greg Pariso MGL at TSRIo Pat Walters, Matt Stahlo Don Bashfordo Guido van Rossum & Python community