ScienceSoftware :: Chemistry :: QSAR :: Vlife QSARPlus™ Suite


	Home


	Shopping Cart


	Contact Us

US TOLL FREE: --- TEL : +49 6136-996090 International: FAX: +49 6136-996096

Search:

Advanced search

Categories

Bioinformatics

Biology

CAD

Chem/Bio Software

Chemistry

Crystallography

Data Analysis

Data Mining

Economics

Engineering Analysis

GeoGraphic

Geology

GeoPhysics

GIS

Graph Plotting

Graphics

Human Resource Plan..

Mathematics

Organizational Develop..

Physics

Programming

Quality Management

Simulation

Statistics

Strategic Planning

TeX Converters

Bestsellers

1. CODESSA PRO™ QSPR/QSAR Software
2. Molcode ToolBox
3. Molegro Data Modeller
4. Vlife QSARPlus™ Suite
5. QSARpro® Suite

Help

ScienceSoftware :: Chemistry :: QSAR :: Vlife QSARPlus™ Suite

Vlife QSARPlus™ Suite #425

Vlife QSAR+^™ handles non-linear relationships with kNN-MFA.
Unique flexibility to couple any of the variable selection method with any model building method to generate QSAR equation. Wide choice of data selection methods for setting test and training sets such as manual selection, random selection and sphere exclusion method. Allows setting of a cross-correlation limit to be considered during variable selection to avoid selection of inter-related independent variables in the model.

Learn more about Vlife QSAR+^™ Suite

QSARPlus^™ included Tools

Vlife Engine™

VLife QSARPlus™ Features

2D QSAR

3D QSAR

Descriptors

QSAR is statistics based methodology, which is accepted and widely used by medicinal chemists world over.

It establishes correlation between chemical structure and biological response. The chemical structure can be represented in terms of several chemical properties that can be either obtained experimentally or can be calculated.

Thus, QSAR is a mathematical model, which can be utilized to quantitatively predict the activity of newly designed molecules.

Advantages

Handle non-linear relationships with kNN-MFA.

Unique flexibility to couple any of the variable selection method with any model building method to generate QSAR equation.

Wide choice of data selection methods for setting test and training sets such as manual selection, random selection and sphere exclusion method.

Allows setting of a cross-correlation limit to be considered during variable selection to avoid selection of inter-related independent variables in the model.

click on image to enlarge

Contribution plot of descriptors selected in a QSAR model of Steroid dataset

Features

Input Data
A set of molecules with known biological response forms the basis to perform QSAR study. It allows to select a folder (directory) which contains set of molecules. A convenient worksheet environment allows easy visualization, selection and manipulation of the QSAR related data. The activity data can be imported/inserted from text, comma separated and excel files in the QSAR worksheet.

Molecular Descriptors Calculation
Using molecular descriptors calculation various 2D/3D molecular structural descriptors (variables) such as physico-chemical, alignment independent and MMFF atom types are calculated .The descriptor calculation is important for finding their correlation with the biological response.

Preliminary Data Analysis
Allows preliminary investigation of data with the help of univariate statistical analysis (minimum, maximum, average, variance, and so on), cross-correlation matrix and graphs to visualize data patterns and inter relationships.

Data Preprocessing
QSARPro allows effective reduction of the data by removing the invariable columns. Data preprocessing is performed by various scaling methods such as no-scaling, mean-centering and auto-scaling. In addition, a variance cut-off option is provided to filter off the noise in the data.

Data Selection

A data selection feature allows selection of dependent and independent variables to be considered in the study. Training and test sets can be created by manual, random and sphere exclusion methods.
Variable Selection Method

Various variable selection procedures like stepwise, simulated annealing, genetic algorithm and so on are provided to select optimal subset of descriptors from available pool of descriptors. It also allows setting of the variable selection method dependent parameters.
Model Building Method

A variety of model building methods ranging from linear like MLR, PCR, PLSR to non-linear like neural networks, k-nearest neighbor are available in QSARPro. The resulting models are shown in terms of descriptors involved in the model and their coefficients. The models can be compared based on corresponding statistical parameters and contribution/fitness plots.
Model Validation

The quality of models obtained is compared based on internal (cross-validation, q2) and external validation (predicted_r2) parameters as well as standard error. In addition, randomization test is also performed to judge quality of QSAR model.
Prediction

The developed model can be utilized to predict activity of newly designed molecules. The developed QSAR model can be coupled with combinatorial library generation utility as a tool for virtual screening.

2D QSAR

2D QSAR establishes correlation between molecular descriptors and biological response. Set of molecules having 2D structure becomes pre-requisite step. 2D QSAR descriptors can then be calculated as physic-chemical, alignment independent, topological, etc. As a result, 2D QSAR leads to identification of various descriptors responsible for activity variation. In addition, it provides a mathematical model, which can be used to quantitatively predict the activity of newly designed molecules.

click on image to enlarge

Advantages

The 2D QSAR model can be used to predict the activity of newly design molecule. The application of 2D QSAR model can further be explored by screening the database of molecules or molecules generated from LeadGrow (Combinatorial library) with easy interface.

Features

Input Data

A set of molecules with known biological response forms the basis to perform 2D-QSAR studies. The activity data can be imported/inserted from text, comma separated and excel files in the QSAR worksheet.

Descriptor Calculation

It offers more than 1000 descriptors for from various categories vis. Topological, alignment independent, atom type count, etc.

Preliminary Data Analysis

It offers preliminary investigation of data by providing univariate statistics, cross-correlation matrix and graphs.

Data Preprocessing

Allows to effectively reduce the data by removing the invariable columns. For data preprocessing various scaling methods such as no-scaling, mean-centering and auto-scaling are provided. In addition, a variance cut-off option is provided to filter off the noise in the data to be considered in the study.

Data Selection

A data selection interface allows selecting a dependent and various independent variables to be considered in the study. For model validation it offers different ways to create train and test set such as manual, random and sphere exclusion algorithm.

Variable Selection Method

To select optimal subset of descriptors from available pool of descriptors various variable selection procedures like stepwise, simulated annealing, genetic algorithm and so on. are provided. It also allows to set the variable selection method dependent parameters like for stepwise Fin, Fout and so on.

Model Building Method

Model Building method offers a variety of model building method ranging from linear like MLR, PCR, PLSR to non-linear like neural networks, k-nearest neighbor. It displays the resulting model descriptors along with its corresponding statistical parameters and contribution/fitness plot in a separated window.

Model Validation

Allows to perform model validation by both internal (cross-validation, q2) and external validation (predicted_r2). In addition, it also allows to perform randomization test.

Results

Shows the contributing grid points in the model and their distribution around set of aligned molecules. It allows to save/retrieve the developed model, which can be used to predict activity of new designed molecule. It offers to use developed model in association with combinatorial library generation utility as a means of virtual screening.

3D QSAR

3D QSAR establishes correlation between local molecular descriptors and biological response. Alignment of a set of molecules becomes pre-requisite step, as local descriptors need to be calculated over common grid around molecules for the analysis. 3D QSAR descriptors can then be calculated as steric and electrostatic interaction energy of a probe atom with each molecule at the grid points. As a result, 3D QSAR leads to identification of various local interacting molecular features responsible for activity variation and hence aid in design of novel potential molecules. In addition, it provides a mathematical model, which can be used to quantitatively predict the activity of newly designed molecules.

Advantages

k-NNMFA is a novel method which utilizes k-nearest neighbor (kNN) method to correlate molecular field descriptors with biological activity.

The predictive ability of kNN-MFA method is better than conventional method of previous generation and MFA models since these methods try to generate linear relationship with the activity whereas kNN is inherently non linear method and hence explains activity trends better than above methods.

click on image to enlarge

Ranges of field values for set of steroid molecules using kNN-MFA

click on image to enlarge

Activity Distribution plot of training and test set using series of substituted 1-(3,3-diphenylpropyl)-piperidinyl amides

Features

Input Data

A set of aligned molecules with known biological response forms the basis to perform 3D-QSAR studies. In addition, it also requires knowledge of active conformation of at least the most active molecule in the set. It allows selecting a folder (directory), which contains set of molecules. The activity data can be imported/inserted from text, comma separated and excel files in the QSAR worksheet.

Molecular Alignment

As the 3D QSAR analysis is conformation and alignment dependent molecular alignment is an essential step. Molecular alignment of the set of molecules can be done by using Align module.

Grid Interaction Energy Descriptors Calculation

The aligned set of molecules is kept in a grid (by default increased 4 angs. in all directions with respective to aligned molecules) to calculate interaction energy fields descriptors such as electrostatic and steric to be correlated with the biological response. It allows setting the grid size/interval and both fields interaction energy cut-off. The values of the calculated descriptors are populated in the QSAR worksheet. A convenient worksheet environment allows easy visualization, selection and manipulation of the QSAR related data.

Preliminary Data Analysis

It offers preliminary investigation of data by providing univariate statistics, cross-correlation matrix and graphs.

Data Preprocessing

Allows to effectively reduce the data by removing the invariable columns. For data preprocessing various scaling methods such as no-scaling, mean-centering and auto-scaling are provided. In addition, a variance cut-off option is provided to filter off the noise in the data to be considered in the study.

Data Selection

A data selection interface allows selecting a dependent and various independent variables to be considered in the study. For model validation it offers different ways to create train and test set such as manual, random and sphere exclusion algorithm.

Variable Selection Method

To select optimal subset of descriptors from available pool of descriptors various variable selection procedures like stepwise, simulated annealing, genetic algorithm and so on. are provided. It also allows to set the variable selection method dependent parameters like for stepwise Fin, Fout and so on.

Model Building Method

Model Building method offers a variety of model building method ranging from linear like MLR, PCR, PLSR to non-linear like neural networks, k-nearest neighbor. It displays the resulting model descriptors along with its corresponding statistical parameters and contribution/fitness plot in a separated window.

Model Validation

Allows to perform model validation by both internal (cross-validation, q2) and external validation (predicted_r2). In addition, it also allows to perform randomization test.

Results

Shows the contributing grid points in the model and their distribution around set of aligned molecules. It allows to save/retrieve the developed model, which can be used to predict activity of new designed molecule. It offers to use developed model in association with combinatorial library generation utility as a means of virtual screening.

Descriptor Calculation

Molecular descriptors can be defined as a numerical representation of chemical information encoded within a molecular structure through mathematical procedure. These descriptors is useful in various area of drug discovery research such as QSAR analysis, chemical database searches, defining molecular finger prints, comparison of molecules, focused library generation and so on. Various types of molecular descriptors can be calculated for a given molecule using different level of molecular information ranging from its molecular formula to 3D geometry.

0D - Derived from molecular formula for example: molecular weight, number and type of atoms and so on.

1D - Requires 1D molecular representation and consists of a list of molecular fragments (For example: functional groups, rings, bonds, substituents and so on).

click on image to enlarge

Spreadsheet shows descriptors calculated from 23 categories of physicochemical properties of Steroid datasets

2D - Requires 2D molecular connectivity information based on how the atoms are bonded in a molecule, for example: total path count & molecular connectivity indices.

3D - There are different types of 3D descriptors for example: electronic, steric, shape and so on like molecular surface area, molecular volume and dipole moment.

Advantages

Allows rapid calculation and population of wide variety of molecular descriptors in a convenient QSAR worksheet.

A graphical representation of the relative distribution of descriptor values in a given set of molecules by distribution/pattern plot is available.

Analysis of descriptors in terms of univariate analysis is available.

Cross-correlation matrix can be obtained to investigate the underlying relationship in/between different group of descriptors.

Features
Descriptor Calculation calculates the following (~1000) molecular descriptors:

Molecular connectivity indices - These are molecular descriptors based on the molecular graph of a molecule. It is calculated for a variety of molecular subgraphs for example: path, cluster, path-cluster and chain of different order. For example path of 1 order means all bonds in the molecule.

Electro-topological indices - These descriptors are calculated for a defined set atom types in a molecule. It is calculated by considering molecular connectivity and electronic environment of each atom type in the molecule.

Thermodynamic - These are properties like logP, molar refractivity based on group additivity.

Electronic - These are the descriptors dependent on 3D geometry and partial charges on atoms of a molecule for example: dipole moment, polarizibilityAHC, polarizibilityAHP and so on.

Steric - These are descriptors for characterizing shape/size of molecule for example: molecular volume, molecular surface area, principal moment of inertia and so on.

Structural - These are descriptors reflecting structural aspect of the molecule for example: molecular weight, count of hydrogen bond acceptor, count of hydrogen bond donor, rotatable bond count and so on.

Molecular surface based - These are descriptors based on molecular electrostatic surface properties for example: MostPositiveElectrostaticPotential, MostNegativeElectrostaticPotential, PositiveElectrostaticSurfaceArea and so on.

Semi-empirical - These are the descriptors calculated by optimizing the molecule using semi-empirical methods like AM1, PM3, CNDO2 etc. for example: heat of formation, dipole moment and so on.

Alignment independent topological and geometrical descriptors - These are descriptors indicating the importance presence of two atom types separated by a given bond distance (topological 2D or geometrical 3D). for example: T_C_O_2 means atom types carbon and oxygen separated by two topological bond distance.

MMFF atom type descriptors - These are the descriptors based on the counts of MMFF atom types.

Allows calculation/selection of descriptors based on group type.

Allows calculation of descriptors for a single molecule for which the result is shown in output window.

Allows calculation of descriptors for a set of molecule by interfacing with QSAR worksheet.

System Requirements

The following lists the supported operating systems and minimum hardware requirements:

Operating Systems	Windows® NT/2000/XP
Optional Operating system:	Windows Vista®
Recommended Hardware Minimum free hard disk space: 1 GB Minimum required memory: 1 GB (2 GB if operating with Windows Vista) Operating system version numbers are the suggested minimum versions.
Graphic Cards	The majority of graphic cards that support hardware-accelerated OpenGL.

Pricing

Special Offer from us
When you purchase Vlife QSAR+^™ Suite, we will be happy to synthesize your order with 5% discount from the standard/list price.

Vlife QSAR+^™ Suite is available with Licensing Term of 1 Year or Licensing Term of 3 Years.

The Pricing for the following License-Areas are available under request.
Feel free to contact us for pricing.

Industry and ‘for – profit’ Research Organizations (Commercial)

Government, Semi-Government and private ‘not-for-profit’ Research Organizations

This software is available with Volume or Site License Option with major discounts.
Contact us for a price quote.

License Area	License Type	1 Year Licensing Term	3 Years Licensing Term
Academic License	Single User License	$ 2,800 Now $2,660	Contact us!

Details

Price: $0.00 (€ 0.00)

Options

select License Type

select Licensing Term

Quantity

Add to cart

Send to friend

Your name: *

Your e-mail: *

Recipient's e-mail: *

Send to friend

Users online

31 unregistered customer(s)

Your cart

	Cart is empty

View cart
Checkout

Authentication

Secure login

Recover password

If Javascript is disabled in your browser click here

News

15-07-2010

New Software on our site:
Voxler^® 2 released!
More details here!

22-04-2010
Molcode Toolbox 2.4 available!
More details here!

05-02-2010
QSARpro® available!
For QSARpro® here!

VLifeMDS™ available!
For VLifeMDS™ here!

ChemXplor™ available!
For ChemXplor™ here!

BioPredicta™ available!
For BioPredicta™ here!

28-01-2010
Morphit available!
For Morphit here!

16-01-2010
PrimerPleX available!
For PrimerPleX here!

17-10-2009
HSC Chemistry® 7 ready!
For HSC 7 here!

Grapher™ 8 released!
For Grapher™ here!

06-08-2009
EMIGMA available!
For EMIGMA here!

GeoTutor IV available!
For GeoTutor here!

QCTool available!
For QCTool here!

Gaussian09 released!
For Gaussian09W here!
For Gaussian09M here!

GaussView5 released!
For GaussView5W here!
For GaussView5M here!

HydroGeoSphere available!
More details here!

DSC Curve Solutions® available!
More details here!

dcsDigitiser Graph Digitizer available!
More details here!

MathType 6.5 German/French Version released!
More details here!

MathDaisy available!
More details here!

New Release of Surfer 9 available!
More details here!

04-04-2009
CODESSA PRO™ available!
More details here!

17-02-2009
VisuMap 2.7 available!
More details here!

17-01-2009
New Releases:

New Software of 'Hyper Family' available!
HyperProtein Pro 1.0! More details here!

FlexPDE Version 6! More details here!

Previous news