2007 Accelrys Science Forum Abstracts

Lennart Nilsson
Karolinska Institute

Ligand Interactions with Nuclear Hormone Receptors
We have used MD simulations to study point mutations in the glucocorticoid receptor (Mol.Endo.19, 1960). Since the ligand is often completely buried in the binding pocket we also used Random Acceleration MD to identify pathways for ligand exit in the retinoic acid receptor (Biophys. J. 91 , 3151).

Darren Fayne
Trinity College Dublin

High-throughput Virtual Screening Drug Discovery - Pipeline to Success
Findings relating to appropriate consideration of database preparation and composition - the implications of input string processing, the distribution of known actives in medicinal chemistry space, the impact of protonation, tautomerisation and multiple conformations on enrichment, and how best to deal with scoring function issues - selection and unbiasing - for optimal performance in the estrogen receptor will be presented. Drawing on many disparate technologies we discuss consolidation of our protocols through Pipeline Pilot, and present on reduction to practice for the identification of active lead compounds.

Esben Friis
Novozymes

Combining Things - An Example of a Friendly Cooperation between Commercial and OpenSource Software

Franck Pitré
Galderma
&
Patricia Gestoso-Souto
Accelrys

Molecular of Non Ionic Surfactant Micelles
A surfactant molecule can be roughly described as a molecule with a hydrophilic head and a hydrophobic tail. Dissolved in water some surfactant molecules gather to make up a spherical aggregate called micelle with the hydrophilic head forming the outer surface of the micelle. These supramolecular aggregates can solubilise hydrophobic molecules in water. Assemblies of micelles form a thermodynamically stable system but incorporation of new molecules and change of temperature can have an impact on these systems Furthermore they can protect hydrophobic molecules from unwanted chemical reactions. But nethertheless molecular interaction can occur between inside and outside through micellar interface leading to chemical reactions. This presentation focuses on molecular simulation of micelles and distribution of molecules with different physicochemical properties. This work is the result of a fruitful collaboration with David Rigby from Accelrys. Atomistic molecular dynamics and coarse-grained dissipative particle dynamics were used to simulated micelles and to gain information on the distribution of the different molecules within the micelles. The numerical data obtained are in accordance with the available experimental data. Important molecular information was obtained which can not
be easily obtained with experimental experiments. New insights of molecular information were produced. Further experimental data and numerical simulation should be generated to confirm these findings.

Julian Willmott
Accelrys

Integrating Accelrys Software in the Scientific Enterprise: Automating Scientific Business Processes and Enabling Scientific Business Intelligence
Science companies revolve around experimentation because experiments produce the raw data for any scientific organisation's most valuable product: intellectual property. Experimentation happens all the way through research and development and is nearly always orchestrated by overriding business processes.
All too commonly, this essential information is hard to search, hard to find or completely lost and expensive experiments end up being repeated. In this talk we will look at some approaches to solving these problems through automating and tracking scientific business processes in disparate areas such as analytical chemistry, biological experimentation and material formulations.

Martin Karplus
Harvard University

Protein Function & Ligand Design
The use of simulations to analyse some fundamental physical characteristics of
proteins will be described and the results obtained for the cellular motor that makes ATP
will be outlined. Examples of the application of simulation methods for combinatorial
ligand design and analysis of ligand activities will be presented.

Frank Brown
Accelrys

The Future of Informatics
Abstract to follow

Catherine Comino
Galderma

Compound Management: From Synthesis to Distribution
Galderma R&D made a large investment in an ensemble of informatics tools to manage compounds from Purchase and Synthesis to Distribution. Accord for Enterprise Informatics (AEI) and Inventory (INV) were chosen by GALDERMA R&D, to manage the Solution library for HTS and the solid compound library for Pharmacology.

In this presentation I will describe:
1. The work flow: from the Accord for Excel (A4XL) data to AEI and INV databases.

2. The specific tools developed for GALDERMA by Accelrys such as:
- For product requests and order tracing
- Client Interfaces to generate customized reports for our clients from the A4XL interface
- Order management by direct transfer of orders to the robots and the updating of the databases after return of the execution files

Ronan Bureau
Université Caen

Identification of Novel Ligands for Human Urotensin-II Receptor through 3D Pharmacophores Virtual Screening of In-house CERMN Chemolibrary
From the structure activity relationship (3D-SAR / Catalyst) studies of the reported ligands, we identified 3D pharmacophores of both non peptide agonists and antagonists of Urotensin II. One of these pharmacophores was used as a query to screen the virtual CERMN chemolibrary which resulted in the identification of novel family of compounds.

Scott Markel
Principal Bioinformatics Architect
Accelrys

Francesco Buonocore
STMicroelectronics

Electronic properties of carbon nanotubes for nanoscale device applications
Interfacing carbon nanotubes either with a metal or with a semiconductor is certainly a most important step in the development of nanoscale electronic devices. Apart from certain important aspects related to the formation of dipoles at the interface, it is well known that the electronic structure alignment of the two materials near the Fermi level is largely determined by the mismatch between the Fermi level of the substrate and the electron affinity/ionization potential of the adsorbed material. Keeping this in mind we have studied the electron affinity (EA) and the ionization potential (IP) of (5,5) and (7,0) single wall carbon nanotubes.

The role played by finite size effects and nanotube termination has been analyzed comparing hydrogen passivated and no passivated nanotubes of different lengths. We show that the EA and the IP are determined by the interplay between the quantum confinement due to the nanotube finite length and the charge accumulation on the edges due to its finiteness. The band structure, EA and IP of carbon nanotube arrays have also been studied. We show that in this case the EA and the IP are also controlled by the array density. Eventually investigations on the highness of the energy barrier between carbon nanotube and metal substrate are presented.

Miguel Gonzalez
Institut Laue-Langevin

Computer Simulation and Neutron Scattering: Adding Value to Experimental Data
Nowadays computer simulations have become a ubiquitous tool in all branches of science. This is particularly true when considering the case of neutron scattering, as the characteristics of the technique and the typical time and length scales covered by modern neutron instruments match the properties that can be easily computed by means of standard Molecular Dynamics (MD) simulations. Therefore ab initio and classical MD techniques can be exploited as a tool to enhance the output of neutron scattering experiments, while at the same time neutron results may serve as a benchmark to validate simulations.

Examples of the combined application of both tools to study the structure and dynamics of very different systems (crystals, fibers, liquids and glasses, porous nanomaterials …) will serve to highlight the natural symbiosis between modeling and neutron experiments and to show the enhanced value that the simulation can contribute to the experiment.

Jocelyne Galy
INSA-Lyon

Molecular Dynamics Simulation of Hybrid Copolymers Based on Methacryl-POSS and Methyl Methacrylate
For the last decade, hybrid organic-inorganic nanocomposites based on the incorporation of polyhedral oligomeric silsesquioxanes or POSSTM into polymeric matrices have received a considerable amount of attention. The goal of the present study is to achieve an understanding of the structure-property relationships in atactic poly(MMA-co-POSS) using atomistic molecular modelling, in comparison with experimental data.

Guillaume Maurin
Université Montpellier

Adsorption of Greenhouse Gases and Hydrocarbons in Flexible Metal Organic Framework Materials : Computational Studies compared to Microcalorimetry Experiments
MOF materials have recently emerged as viable candidate adsorbents for a number of different gases, for instance, carbon dioxide and methane, which are potent greenhouse gases that are implicated in existing hydrogen production processes. With a future world economy based on hydrogen fuels imminent, coupled with the recent world commitment to the Kyoto Agreement, there is a need for materials to efficiently capture and store these types of gas molecules. The aims of this present project is the estimation of the breathing phenomena in flexible hybrid solids in the presence of either apolar (alkanes) or polar (carbon dioxide) vapors using Microcalorimetry measurements , (ii) the understanding of this phenomenon, through various computational tools including Grand Canonical Monte Carlo (Sorption) and Density Functional Theory (Dmol3) simulations. The final goal of this project will be to reach a “design” of selective MOFs sorbents for their use in fluid storage (alkanes, alcohols) and separation processes, with a low-cost regeneration, what is of strategic interest in the domain of energy and environment.

In this work, we will present an exploration of the adsorption mechanism of CO2 and different alkanes (from C1 to C4) using complementary Monte Carlo and Density Functional Theory techniques in MOF materials taken from the MIL (Materials of the Institute Lavoisier) series, containing different metal centres. This strategy has been succesfully applied to various MIL systems [1-3]. We predict the adsorption isotherms and enthalpies of adsorption and probe the possible adsorption sites of the different adsorbates in the MIL-53 (Al,Cr,Fe) and MIL-47 (V) materials. These simulations are directly compared to microcalorimetry and in situ X-ray diffraction data.

Stephen Todd
Accelrys

Future Developments in Materials Studio 4.2
This presentation will highlight some of the new functionality in Materials Studio 4.2. This will include new functionality exposed through the MaterialsScript scripting language, new modules such as the Gaussian Interface and Adsorption Locator, and recent developments from the Nanotechnology Consortium.

Béatrice Charton
Crihan

A Regional Scientific Software Platform
The Centre de Ressources Informatiques de Haute-Normandie (CRIHAN) is a regional non-profit organization providing networking and supercomputing resources to public research laboratories and to the industry. CRIHAN is especially in charge of the regional network for research, health and education and provides computer means and high-end software for chemists and biologists. Both communities have been in the last few years very active and it has therefore been decided to set up a common software platform, accessible to every public lab, to provide a fairly large set of up to date scientific software and to reduce the costs of maintenance.

Martin Karplus
Harvard University

Protein Function & Ligand Design
The use of simulations to analyse some fundamental physical characteristics of
proteins will be described and the results obtained for the cellular motor that makes ATP
will be outlined. Examples of the application of simulation methods for combinatorial
ligand design and analysis of ligand activities will be presented.

Robert Klein
BayerCropScience

Optimised Docking Protocol for Structure-based Virtual Screening
Docking protocols and scoring functions for structure-based virtual screening and ligand design are aiming at identifying the preferred ligand orientations and conformations (poses) within a protein and estimating the protein-ligand binding affinities. While classical force field approaches perform quite well in pose prediction, they seem to be less suited for assessing binding energies. We present a virtual screening protocol which makes use of a pose generator in conjunction with a classical force field approach to identify reasonable poses and a novel scoring function to adequately account for solvation effects.

Tim Clark
CEPOS Insilico

Surface-based in silico Techniques
NDDO-based semiempirical MO-techniques provide one-electron properties of molecules very economically. Local properties derived from semiempirical wavefunctions provide an information-rich description of molecules that is effective for QSAR, QSPR and prediction of reactivity. Some recent examples based on ParaSurf'07 will be described.

Florian Von Wrochem
Sony

The Role of the Metal-Molecule Interface in Molecular Electronics
The potential of molecular electronics for future device application critically depends on the formation of a stable and defined link between organic molecules and metallic contacts. Here we present investigations on dithiocarbamates used as an alternative connecting group, showing interesting properties regarding stability and electrical conductance. The structure of dithiocarbamate based self-assembled monolayers is elucidated using photoelectron spectroscopy and DFT calculations, whereas their electrical properties are studied by conductivity measurements in networks of interlinked nanoparticles. Some general procedures for the investigation of self-assembled monolayers by molecular modeling, in particular using DFT methods, are described, giving insight into monolayer structure and electronic properties.

Bernhard Wunderle
Fraunhofer IZM

Moisture Diffusion in Epoxy Resins using MD Simulation and Experiment
Moisture represents a potential risk for the reliability of polymer interfaces within electronic packages as it changes their thermo-mechanical properties. This paper focuses on the diffusion of moisture in epoxy resins in experiment and molecular dynamics simulation as a function of molecular structure and external boundary conditions.

Richard Gilbert
e2v Biosensors

Versatile Biosensor Labels
Over the past two decades biosensors have increasingly made an impact in the life sciences and medical markets, and are beginning to be used as mainstream research tools in many laboratories. Biosensors are devices that use biological or chemical reactions to detect analytes in a sample, typically monitoring the levels of specific proteins or nucleic acids in biological fluids such as blood or cell extracts. There are many different biosensor detection technologies, each of which bring their own benefits and restrictions, and which have typically required the development of specific chemical labelling systems in order to make best use of the sensor technologies.

Using Materials Studio, e2v has developed a generic labelling chemistry which can be applied to a variety of detection technologies. A single, generic chemical framework has been developed which accommodates sensor technologies as diverse as electrochemistry, fluorescence, chemiluminescence, electrochemiluminescence and Raman spectroscopy, and therefore addresses all of the major detector types in current use.

Thomas Weber
ETH Zurich

A DFT study of the hydrodesulfurization of dibenzothiophene
Hydrodesulfurization (HDS) of crude oil belongs to the most important applications of heterogeneous catalysis worldwide. The HDS of dibenzothiophene on a NiMo18 sulfur cluster is investigated by means of DFT calculations using the DMol3 code. The calculations focus on the direct desulfurization pathway where dibenzothiophene reacts with H2 to biphenyl and H2S. Critical steps with high activation energies are breaking of the two C-S bonds of the DBT molecule (ca. 32 kcal/mol), the activation of H2 by heterolytic dissociation and the removal of the formed biphenyl molecule from the catalyst surface (17 kcal/mol, respectively). All other reaction steps have activation energies of £ 10 kcal/mol.

Gerhard Goldbeck-Wood
Accelrys

Future Developments in Materials Studio 4.2
This presentation will highlight some of the new functionality in Materials Studio 4.2. This will include new functionality exposed through the MaterialsScript scripting language, new modules such as the Gaussian Interface and Adsorption Locator, and recent developments from the Nanotechnology Consortium.