MOLPHARM: CHEMOINFORMATICS TOOL WITH 3D INTERFACE AND .NET SUPPORT

Developed by Sushil Ronghe

As a part of M.S. project



Molecular modeling software enabled the scientist to visualize the compounds they worked with as early as 1970. Although the early versions only displayed black and white pictures of the atoms within small compounds, it was a powerful tool for developing a mental model of compound and it’s property.The technology advances in computer science result in number of free and commercial software that gives a very sophisticated representation of the chemistry. OpenGL and java3D are the graphics libraries mostly used by visualization program. Both of these libraries are free and can be obtain from their official web site.


Java3D official web site
OpenGL Web site

The visualization in molecular modeling comes under two heading molecular structure visualization and molecular property visualization. The fundamental approach to display molecules and their properties is same for all molecular modeling software. The structural information is the key requirement for the visualization of molecule.


Structural Information Files

Structural information of molecule can be obtained from experimental technique like crystallography/NMR etc. Different file formats have been developed since 1970 to represent the structural information,parallel to that MDL molfile become a defecto standard. Molpharm is chemoinformatics tool with number of functionality,like chemical file input/output,database reading/writting,visualization,structure analysis,and ADME property calculation etc. As far the chemical structure file input is concern molpharm support fallowing files


File Format
File NameExtension
Protein Data Bank*.PDB
Structural Data File*.SD
MDL Mol File*.MOL
Sybyl Mol File*.MOL/*.MOL2
Thinker MM2 Input*.XYZ
Thinker MM3 Input*.XYZ
Sybyl 2 file format*.SM2
Sybyl file format*.SML
Alchemy file format*.ALC
Cartesian Coordinate*.CC1
Cartesian Coordinate*.CC2
Connection Table*.CT
EXtented MarkUp Language*.XML
PDB Mark Up language*.PDBML

Molpharm can read 11 file format listed above,Molpharm readers are developed from a scratch.


Protein Models

Some large figures are kept in gallary

Structural models are well known in the field of the CAMD(computer aided molecular design).In molpharm visualization of the small molecule as well as of the protein molecule is the initial task.The color scheme avilable are CPK temperature,Monochrome and group.The color information and the models are made with the reference from popular molecular visualization software RASMOL. The PDBFileReader program in molpharm is coded with lot of functionality to satisfy the need of the future developer.


C alfa Trace/Secondary Structure

This is single line ribbon representation of the protein back bone.The first model is of protein ID 1H1H and next is of protein 1PEN The final version will contain the more sophisticated view of the secondary structure. Mouse interaction is also associated with the residue in the protein and so the user can select one residue at a time. Color of the each residue can't be change at this time but it can be possible.The ribbon are BSpline curves. Secondary structure utility is also coded from the scratch.


Appearance

Visualization comes in reference of seeing.It also signifies 'intelligible' or 'intuitive'.For viewer the appearance of the object do everything.Different color values are used for the each element.The color values are taken from the literature .The appearance of the atom is combination of the color values like ambient colorDiffuse color,specular color,shine and object color etc. These color values are then mapped to the color values in java3D.


ATOM COLOR
AtomColor
Carbonlight Gray
OxygenRed
Sulpheryellow
Nitrogenblue
HydrogenCyan
PhosphorosOrange



Small Molecular Representation

The small molecule representation has a fallowing model for visualization Ball and stick ,stick ,Wire frame,and space fill .The figure above is showing all these model for aspirin molecule (Figure is created in molpharm).Behind the screen there is lot of information being obtained for the molecule on the screen.Like ring-perception,atom type identification etc. This information can also be made visible but it is not of any use for the user.In documentation of the program all methods and member are given which can be used to reprogram molpharm to change the view of the software or to print some of the information which the user want's specifically.


Chemoinformatics libraries

Mopharm provides a nice interface for opensource projects like CDK ,JOELib and Biojava. At Present Descriptor calculation and Matching API for substructure search is used from CDK.


PHARMACOPHORES

Figures are kept in gallary

Pharmacophoric features include all those binding related structural or chemical properties of chemical compounds that are thought to be responsible for a specific pharmacological action. Chemical features taken into account usually include hydrogen bond donor/acceptor, charge, hydrophobicity and aromacity. In the present approach individual atoms are labeled with pharmacophore properties. These labeled atoms are often referred to as pharmacophore points. There are numerous methods available to perceive pharmacophore points in chemical compounds. THe method we have adopted is the knowledge- or rule-based approaches.Ring perception(SSSR/ESSR) substructure search are most common of them. Molpharm only provide the identification of the pharmacophoric points in the structure. THe pharmacophore points that can be identified are HD,HA,Hydrophobic center,Cation and anion. The work in this regard is in progress.


Functions Available

Present functionality of the molpharm are describe below.My aim is to keep the utility helpfull to the user and made a beautiful interface like other popular software.


File Format
FunctionWork
ChemFileReading and writting the Files
CalculateCalculate over 30 Molecular descriptor
Distace TableDistance,Torsion and Angle calculation
Search Substructure Search and MCSS with Graphical Analyser
Database utility/SD EngineThis is easy way to prepare/manage the chemical databank
ADMEPresent descriptor include logS,logP
AnalyserGraphical analysis of structure model/color

AlogP as well as the logS is calculated from web service provided by
Virtual Computational Chemistry laboratory


SD Engine

SD engine provides the interface for SD file to be used as a usual files. You can download any sd file directly to molpharm because of .net support.When the file is being processed a nice progress bar(implemented in jdk 1.6) will tell you how much time will it take to complete.Once molpharm finished with the file a SD assembly will prompt you about the properties found in the sd file and those which you want to be in the tabel.After the selection of the properties tabel will pop up with the serial number and the property values.On the screen you will see the first molecule.This is the default .As soon as you clicked on the row the molecule will be visible in the window.You can filter these molecule as well as calculate the property that molpharm support.Another use of this engine is preparation of the chemical database.Give the directory in which you have your molecules,the file format should be the one which are enlist at the top of the page.Molpharm will convert them in the tabel format and now you can calculate the property that molpharm support and make a databank. To summerize, SD Engine can be used to add/remove extra field to the present database and create a new database with the filed that molpharm support.

All the latest things added to SD engine is Substructure search,Fingerprinting calculation and MCSS search.This utility is found to be very usefull.The future work of molpharm is concentrated on similarity search and pharmacophore mapping as well as combinatorial exploration. These all modules will be inherited from SD ENGINE.
Note ::A brief tutorial will be available soon.

Substructure Search

Substructure search is one of the important task in molecule modelling.It is used for various purpose like persiving the chemical environment in the molecule for pharmacophore generation or combinatorial purpose.Substructure searching has wide functionality explaining that is certainly out of the scope for this document.Substructure search in molpharm is very simple and it is also accompanied with the graphical analyser.A simple interface is developed where user can input a smart expression which is then searched againest the structure and those atom which are matching with the query is shown with the green Box. To explain this further here is one example,pioglitazone: this drug belong to the distinct class of glitazones acting on PPAR-Gamma receptor. The substructure that is responsible for the activity of this class of drug is the [C=O]-N-[C=O] where every funtional groupinteract with the respective amino acid on PPAR receptor.The smart expression for the functional group is [OX1]=[CX3]~[NX3]~[CX3]=[OX1] which has given one hit on the pioglitazone and the atoms which are hits for this query are shown in the green box.


Maximum Common substructure search

Consider the two structures in figure above.The left one is target and right one is query. We will perform the MCSS between these two compounds.Open the query molecule in Molpharm and go to "Search3D ==> Remember" this will remember the structure in molpharm memory and use it as a query structure as shown in the figure below. (Molpharm memory will keep the reference of this query structure until the remember menu is click again) Then open the target molecule in molpharm and go to "Search3D==>MCSS" the resultant MCSS found will then be shown in the green boxes.


Future

Complete Structure base Drug Design Java tool
In future combinatorial exploration,similarity search and pharmacophore mapping are the main thing that is going to be incorporated.The aim behind this project is to provide the complete SBDD tool.This site is devoted to all the devloper who are coding for chemisty.Please post your stuff and made it available to every body.

Free Software, Free Society


Home