Showing papers by "James C. Phillips published in 2000"

Speech/gesture interface to a visual-computing environment

Abstract: We developed a speech/gesture interface that uses visual hand-gesture analysis and speech recognition to control a 3D display in VMD, a virtual environment for structural biology. The reason we used a particular virtual environment context was to set the necessary constraints to make our analysis robust and to develop a command language that optimally combines speech and gesture inputs. Our interface uses: automatic speech recognition (ASR), aided by a microphone, to recognize voice commands; two strategically positioned cameras to detect hand gestures; and automatic gesture recognition (AGR), a set of computer vision techniques to interpret those hand gestures. The computer vision algorithms can extract the user's hand from the background, detect different finger positions, and distinguish meaningful gestures from unintentional hand movements. Our main goal was to simplify model manipulation and rendering to make biomolecular modeling more playful. Researchers can explore variations of their model and concentrate on biomolecular aspects of their task without undue distraction by computational aspects. They can view simulations of molecular dynamics, play with different combinations of molecular structures, and better understand the molecules' important properties. A potential benefit, for example, might be reducing the time to discover new compounds for new drugs.

BioCoRE: A Collaboratory for Structural Biology

Abstract: Modern computational structural biology requires scientists to employ a wide range of tools and techniques to solve complex problems while keeping accurate records of research activities. Additional complications are introduced by the need to effectively engage in interdisciplinary collaborations with geographically dispersed colleagues. The software BioCoRE, a collaborative research environment for molecular modeling and simulations, addresses these challenges. Initial design work has led to a web-based architecture focused on four primary interface paradigms: a workbench allows diverse computational tools to be applied to the problem at hand in a consistent manner, a notebook automates recording of research activities, electronic conferences held with collaborators can be saved and replayed, and multi-author documents can be prepared in a crossplatform revision control system. When complete, it is expected that the BioCoRE meta-application will drastically reduce the effort and expense presently associated with structural biology distance collaborations. INTRODUCTION Structural biology investigates the molecular basis of life in its healthy and diseased states. Initially through x-ray diffraction (Perutz et al., 1960; Kendrew et al., 1960), later through NMR (Wuethrich, 1986), and lately also through electron microscopy (Nogales et al., 1998), ever more medically relevant and larger biomolecular structures have been discovered. The blossoming field of genomics rivals structural biology with an outpouring of genomes. The vast amount of information available and the complexity of the information units, gene sequences and protein structures, have made computers indispensable tools in biomedical research to an extent which in 1991 was characterized as a paradigm shift by W. Gilbert (Gilbert, 1991). Today about 200 different programs serve researchers well, in particular those coupled with web-based tools such as the Biology Workbench∗, PROPSEARCH† (Hobohm and Sander, 1995), and BLAST ‡ (Altschul et al., 1990). Programs for structural biology have been mostly single-user oriented and not integrated with webbased tools. Here we suggest to develop a “Biological Collaborative Research Environment” (BioCoRE), an integrated, web-based, and tool-oriented ∗URL: http://biology.ncsa.uiuc.edu/ †URL: http://www.embl-heidelberg.de/prs.html ‡URL: http://www.ncbi.nlm.nih.gov/BLAST/ computing and communication system for biomolecular modeling and simulation. Structural biology, similar to its molecular biology parent, traditionally existed in small, independent laboratories. The need for expensive instrumentation (e.g. x-ray sources), the scale of tasks such as sequencing the human genome, the sharing of large data bases, the requirement of broad expertise and of cutting-edge technology have all led to many trans-mural groups ranging from collaborations between two laboratories to national consortia. Fortunately, this development coincides with the great performance increase of the US research network so that communication, data sharing, and joint use of instrumentation and computing facilities can be realized on the necessary scale (Schooler, 1996; Kouzes et al., 1996; Shortliffe et al., 1996; Clutter, 1996; Finholt and Olson, 1997). The wide bandwidth network connects the structural biology community today to massively parallel powerful computers, “pumping data back to their local site for immediate visualization” (UCAID, 1997). An adequate use of this resource for analysis and modeling of increasingly complex structures or for real time interactive modeling, requires a new type of web-based modeling which exploits distributed computing as well as a unification of graphics and molecular dynamics (MD) simulation. Collaboratory software tools Collaborative projects in structural biology demand new software tools which can be classified into four categories: • distributed resource utilization: transparent allocation of resources, sharing of data, information and disk space • distributed simulation and visualization: remote simulation and visualization control through web-interfaces and interactive molecular dynamics • analysis and postprocessing: web-based analysis tools interfaces, monitoring capabilities, reporting and publication tools • other web-based collaboratory tools for: communication (including audio/video capabilities), mentoring, record keeping, and program repositories Currently, no software package which encompasses all these properties exists. Implementing the aforementioned capabilities in a single application will significantly advance the research environment in structural biology. BioCoRE design supports four basic types of activities pertinent to most research projects: utilizing a wide range of computational tools, keeping records, communicating with collaborators, and writing articles and reports. This functionality will be implemented in four main interfaces of BioCoRE, called Workbench, Notebook, Conferences, and Documents (Fig. 1). The Workbench interface of BioCoRE includes features for controlling molecular modeling, simulation, and bioinformatics tools with convenient and uniform access to collaboratory data. The Notebook interface furnishes the tools for logging, locating, and reviewing methodology, data, results, and annotations related to the ongoing projects. Scientists are able to discuss their research in real time or timedelayed sessions via the Conferences interface, which spawns software for teleconferencing and synchronized visualization of shared data at distant sites. The Documents interface of BioCoRE supplies collaborators with a convenient front end for preparing multi-author documents for publication. BioCoRE is a general web-based collaboratory tool that leads to accelerated development and dissemination of basic biomedical knowledge. This networkcentered meta-application improves the collaboration between biomedical researchers located at either the same institution or at geographically distant places. It facilitates the transparent use of and communication between existing programs, tools, and databases. BioCoRE allows researchers to share information and resources. Scientists interact in both synchronous and asynchronous fashion with each other or with the modeling tools via a common infrastructure. BioCoRE enables scientists to initiate new collaborations through its communication interface and reduces the need for travel between the participating research groups. EXISTING COMPONENTS OF

Scalable Molecular Dynamics for Large Biomolecular Systems

Abstract: We present an optimized parallelization scheme for molecular dynamics simulations of large biomolecular systems, implemented in the production-quality molecular dynamics program NAMD. With an object-based hybrid force and spatial decomposition scheme, and an aggressive measurement-based predictive load-balancing framework, we have attained speeds and speedups that are much higher than any reported in literature so far.The paper first summarizes the broad methodology we are pursuing, and the basic parallelization scheme we used. It then describes the optimizations that were instrumental in increasing performance, and presents performance results on benchmark simulations.

Scalable molecular dynamics for large biomolecular systems

Abstract: We present an optimized parallelization scheme for molecular dynamics simulations of large biomolecular systems, implemented in the production-quality molecular dynamics program NAMD. With an object-based hybrid force and spatial decomposition scheme, and an aggressive measurement-based predictive load balancing framework, we have attained speeds and speedups that are much higher than any reported in literature so far. The paper first summarizes the broad methodology we are pursuing, and the basic parallelization scheme we used. It then describes the optimizations that were instrumental in increasing performance, and presents performance results on benchmark simulations.