Thomas Daniel Crawford, Ph.D. - US grants

T. Daniel Crawford is supported by a CAREER grant from the Theoretical and Computational Chemistry Program to develop and apply accurate quantum chemical methods to the chiroptical properties of large molecules. He will design practical computational tools for the determination of the absolute configurations for newly synthesized chrial molecules and for insight into optical activity. He will also develop new algorithms to overcome the polynomial scaling of coupled cluster theory. In his educational activities, he will develop an integrated computational chemistry program across the undergraduate chemistry curriculum, involving computational experiments for the lab courses.

Optical properties of large chiral molecules are of interest to natural product research in order to manipulate and synthesize stereochemically pure chiral drugs. The ultimate goal of this research program is to develop a computationally efficient and accuate method for calculating these properties from first princilpes. The goals of this educational program are not only to teach students the fundamentals of computational chemistry and for what chemical problems computational techniques may be the preferred analytical tool, but also to help them understand why and when such tools might fail. The lab projects will be collected into a computational chemistry handbook to be published in print and on the web for use by other chemistry departments.

T. Daniel Crawford of Virginia Polytechnic Institute and State University is supported by the Theoretical and Computational Chemistry Program to develop methods to provide accurate predition of the optical properties of large chiral molecules. The methods are based on a gauge- and origin-invariant reformulation of coupled cluster theory. This approach is systematically testable and convergent. A reduced-scaling implementation of this method is under development that will allow calculations of molecules with up to several dozen atoms. Applications include calculations of optical rotation and circular dichroism spectra, and direct comparison with experiment. This work is having a broader impact in that the methods developed in this work are freely available through the PSI3 program suite. Success in implementing the proposed methods will provide valuable insight in assigning absolute stereochemistry of natural products. Educational efforts include training the next generation of scientists in a set of skills related to computational science. The educational laboratory program is being expanded to include computational laboratory experiments at the undergraduate level.

With this award from the Chemistry Research Instrumentation and Facilities: Multi User Program (CRIF:MU), the Department of Chemistry at the Virginia Polytechnic Institute and State University will acquire a large-scale computing cluster (96 nodes) for the study of reduced-scaling electronic structure models for chiral molecules, first-principle predictions for properties of molecules and materials, improved semiempirical Hamiltonians for reaction dynamics, biophysics of lipid bilayer membranes as well as computational studies related to surface chemistry and catalysis.

A cluster of fast, modern computer workstations is vital to serving the computing needs of active research departments. Such a "computer network" also serves as a development environment for new theoretical codes and algorithms, provides state-of-the-art graphics and visualization facilities, and supports research in state-of-the-art applications of parallel processing. The computing resource will permit research, research training, and research outreach opportunities.

T. Daniel Crawford of Virginia Polytechnic Institute and State University is supported by the Chemical Theory, Models and Computational Methods Program in Chemistry and the Office of Cyberinfrastructure to develop methods that provide accurate predictions of the optical properties of chiral molecules in solution. The methods are based on coupled cluster linear response theory extended to incorporate solvent perturbations via both explicit and implicit models. This approach is systematically testable and convergent, benefitting from reduced-scaling implementations under development in the Crawford group that allow calculations involving many dozens of atoms. Applications include calculations of solution-phase optical rotation and circular dichroism spectra, and direct comparison with experiment.


Results of this research include new chemical models that are embodied in sustainable software made freely available through the open-source PSI program package. The research provides valuable insight into the assignment of the absolute stereochemistry of natural products, as well as into the fundamental connections between molecular structure and the optical properties of chiral molecules. Educational efforts are directed towards training the next generation of computational scientists and the continued development of undergraduate computational laboratory experiments.

This award will enable an international collaboration for software development in computational chemistry between researchers in the United States and the United Kingdom. This international collaboration arose from a joint NSF/EPSRC workshop on Software Development for Grand Challenges in the Chemical Sciences held in Oxford, United Kingdom in June 2011. This award will fund planning and community building workshops, and exchange visits to catalyze projects.

The workshops will lay the foundation for the development of new community software that will impact all quantum chemistry programs. The resulting improvements will subsequently benefit the array of computational chemistry and physics software (e.g., solid-state physics and molecular dynamics codes) that depend on quantum chemistry models, and will ultimately provide a path to solving grand challenge problems such as large-scale ab initio simulations of metallic systems, the spectroscopy of solvated systems, or photoactive biological processes.

Quantum chemistry can provide highly accurate results for arbitrary molecular systems, making it a vital component in many different disciplines such as materials science, biology, physics, chemical engineering, mechanical engineering, environmental science, geology, and others. It is particularly critical in the rational design of drugs, catalysts, organic electronics, nanostructured materials, and other designed materials. Because of their steep computational costs, quantum chemistry codes must exploit parallel computing and must constantly adapt to rapidly changing high performance computing technologies. This creates a significant barrier for the adoption of new technologies into quantum chemistry codes. Our project involves the development of a parallel, highly reusable library for advanced numerical approximations in quantum chemistry. This will be the first unified library of such techniques, designed for high performance and also reusability by independent research groups. The PANACHE (PArallel Numerical Approximations in CHemistry Engine) library will fill this need. To maximize its impact, PANACHE is being designed to be used by multiple quantum chemistry software packages. PANACHE dramatically speeds up quantum computations, making it much easier to gain insight into a wide array of problems, from studies of reaction mechanisms in catalysis to the design of improved organic photoelectronic devices.

Our highly interdisciplinary project (involving two theoretical chemists and one computational scientist as co-PI?s) provides excellent opportunities for training graduate students and postdocs in the areas of numerical methods, high-performance computing, quantum mechanics, and computational chemistry. Computer code resulting from this project will be released as freely-available open-source software, enabling its use with any other software package. Workshops on the new software will be held to introduce these new tools to other software developers, and online training material and graduate course material will be developed to improve education in the use of numerical methods in computational science and quantum chemistry.

This award pertains to the Software Infrastructure for Sustained Innovation (SI2) solicitation.

Computational chemistry is one of the pillars of computational science, and thus its impact reaches well beyond chemistry into biomolecular and polymer physics, materials science, and condensed-matter physics. Over its long history, which stretches back more than half a century to the dawn of computation, computational chemistry has achieved a much-deserved status as a full partner with experiment in scientific discovery, yielding simulations of such high accuracy that its predictions of a variety of molecular properties may be considered "computational experiments," often with greater reliability than laboratory measurements for many chemical properties.

The history of computational chemistry endows the field not only with great experience, but also with a legacy of diverse and complex code stacks. Many molecular dynamics and quantum chemistry programs involve hundreds of thousands to even a million lines of hand-written code in a variety of languages, including Fortran-77, Fortran-90, C, and C++. While this complexity has arisen naturally from the intricacy of the problems these programs were designed to solve, it also presents a crucial obstacle to the long-term sustainability and extension of the software on ever-changing high-performance computing hardware.

The goal of the S2I2C2M2 will be to overcome these obstacles of both algorithms and culture and change the fundamental nature of computational chemistry software development. In the year-long Conceptualization Phase, S2I2C2M2 will bring together an interdisciplinary team of computational chemists, computer scientists, applied mathematicians, and computer engineers to attack the fundamental problems of software complexity and education. Three working groups will focus on the key areas of portable parallel infrastructure, general-purpose tensor algebra algorithms, and protocols for information exchange and code interoperability. In addition, experts from the S2I2C2M2 team will participate in an inaugural summer school on software development for computational chemistry.

Computational chemistry has become ubiquitous throughout the fields of science and engineering. Among the many uses of computational chemistry codes are to aid in the design of new materials with specific properties, to understand protein folding dynamics related to the human body, to understand the mechanisms of enzyme catalysis to produce biofuels, and to identify and characterize gaseous species in the atmosphere. Because computational chemistry simulations can require large amounts of computer, memory and disk space, one focus of this research is to reduce the demands on computer resources by exploiting methods that can efficiently fragment large molecules into smaller pieces and at the same time take advantage of computers that have many thousands of computer cores. The principal investigators are developers of several computational chemistry programs, each of which has unique functionalities that have taken multiple person-years to develop and implement. In order to minimize further development efforts and to maximize the utility of the unique features, a second key focus of this research will be to develop the ability of each program to access the unique features and data of the other programs. A major bottleneck in the effort to construct high performance computers with more and more compute cores is the power that is required to drive such systems. One way that the research team will address the power issue is to explore the utility of low power architectures, such as graphical processing units for computational chemistry calculations. All of the newly developed codes will be made available to the user community by web download at no cost.

This project presents an integrated computational science approach to very high quality electronic structure and dynamics calculations that will (a) be broadly accessible to both the development and applications communities, (b) have the capability to address problems of great interest, such as the properties of liquids and solvent effects, and photochemical/photobiological dynamics with high accuracy, (c) provide interoperability and sustainability into the foreseeable future, and (d) solve bottlenecks including power consumption using accelerators. A primary goal is to provide to the broad community new, accurate approaches that may be easily used, with a clear path forward to further code improvement and development. As part of the proposed effort, in addition to the usual outlets of journal articles and public presentations, the investigators will organize workshops at prominent national meetings, so that the expertise and software developed will be available to as broad a group of users as possible. All of the developed codes will be available on the web for easy downloads. The proposed research will develop new paradigms for interoperability, with a focus on the highly popular program suites GAMESS, NWChem, PSI4 and the AIMS dynamics code. Also included will be a cloud-based client-server model, a common quantum chemistry driver and novel data management approaches. The integration of the codes will be accomplished by making use of the combined expertise of the PIs in developing interoperable methods and data interfaces in computational science. In addition, several new methods will be developed, including novel explicit (R12) correlation methods that will be integrated with the most accurate levels of theory: multi-reference and the most accurate and novel coupled cluster methods, as well as the derivation and implementation of analytic derivatives. Applicability to large molecular systems will be made feasible by drawing upon novel fragmentation methods that scale nearly perfectly to the petascale.

T. Daniel Crawford of Virginia Tech University is supported by an award from the Chemical Theory; Models and Computational Methods program in the Chemistry Division to develop theoretical and computational methods for the accurate prediction of the optical properties of chiral molecules in solution remains a challenging task for computational chemistry. A chiral molecule is one in which there are two different versions that are identical in composition but which cannot be superimposed, much like human hands. The "left-handed" and "right-handed" molecules are mirror images of one another. Such molecules are optically active, they can rotate plane-polarized light either clockwise or counter-clockwise. Chiral molecules play an important role in many branches of chemistry, including the chemistry that governs the functioning of living beings. Crawford and his research group focus on the development of robust and efficient methods for computing the special "chiroptical" properties of these molecules, with the overarching goal of providing practical and reliable tools for assigning absolute stereochemical configurations of chiral compounds. These studies are invaluable in the elucidating the intimate relationship between molecular structure and chiroptical response and thus may provide the chemical community with a deeper understanding of the nature of optical activity.

The best approach for modeling the intimidating complexity of solvent effects on molecular properties remains elusive, especially for chiroptical properties such as optical rotation (OR), circular dichroism (CD), vibrational Raman optical activity (ROA), and circularly polarized luminescence (CPL), which exhibit an exquisite sensitivity to environmental effects. This project focuses on the development of robust and efficient tools for the high accuracy modeling of chiroptical properties in the liquid phase, including implicit, explicit, and hybrid approaches. Implicit models will be developed to approximate the bulk solute-solvent interaction using both shape- and density- based cavities coupled with potentials derived self-consistently. Explicit solvation effects will be accounted for using a new hybrid QM/MM scheme, including a novel wave-function-embedded periodic approach. Efficient implementation of these models will rely heavily on emerging reduced-scaling methods, particularly within the coupled cluster linear-response approach for non-resonant optical activity.

The Molecular Sciences Software Institute (MolSSI) will become a focus of scientific research, education and scientific collaboration for the worldwide community of computational molecular scientists. The MolSSI aims to reach these goals by engaging the computational molecular science community in multiple ways to remove barriers between innovations that often occur in small single-researcher groups and the implementation of these ideas in software that is used in the production of science by the entire community. Thus, great ideas will not languish in the "just get the science right" mode, but be incorporated into usable software for the wider community to enable bigger and better molecular science. The MolSSI will catalyze significant advances in software infrastructure, education, standards, and best-practices. These advances are critical because they are needed to address the next set of grand challenges in molecular science. Activities catalyzed by the Institute will improve the interoperability of the software used by the community, make easier the use of this software on the varied and heterogenous computing architectures that currently exist, enable greater scalability of existing and emerging theoretical models, as well as substantially improving the training of molecular-science students in software design and engineering. Through the range of outreach efforts by its multiple institutions, the MolSSI will engage the community to increase the diversity of its workforce by more effectively attracting and retaining students and faculty from underrepresented groups. All of these endeavors will result in fundamentally and dramatically improved molecular science software and its usage, that will reduce or eliminate the current delays - often by years - in the practical realization of theoretical innovations. Ultimately, the Institute will enable computational scientists to more easily navigate future disruptive transitions in computing technology, and most importantly, tackle problems that are orders of magnitude larger and more complex than those currently within their grasp and to realize new, more ambitious scientific objectives. This will accelerate the translation of basic science into new technologies essential to the vitality of the economy and environment, and to compete globally with Europe, Japan, and other countries that are making aggressive investments in advanced cyber-infrastructure.

The MolSSI aims to reach these goals by engaging the computational molecular science community in multiple ways to remove barriers between innovations that often occur in small single- principle investigator groups and the implementation of these ideas in software that is used in the production of science by the entire community. The MolSSI will create a sustainable Molecular Sciences Consortium that will develop use cases and standards for code and data sharing across the software ecosystem and become a focus of scientific research, education and scientific collaboration for the worldwide community of computational molecular scientists. The Institute will create an interdisciplinary team of Software Scientists who will help develop software frameworks, interact with community code developers, collaborate with partners in cyber-infrastructure, form mutually productive coalitions with industry, government labs, and international efforts, and ultimately serve as future experts and leaders. In addition, the Institute will support and mentor a cohort of Software Fellows actively developing code infrastructure in research groups across the U.S., and, in turn, they will engage in MolSSI outreach and education activities within the larger molecular science community. Through a range of multi-institutional outreach efforts, the Institute will engage the community to increase the diversity of its workforce by more effectively attracting and retaining students and faculty from underrepresented groups. The Institute will educate the next generation of software developers by providing workshops, summer schools, on-line forums, and a Professional Master's program in molecular simulation and software engineering. MolSSI will be guided by an internal Board of Directors and an external Science and Software Advisory Board, both comprised of leaders in the field, who will work together with the Software Scientists and Fellows to establish the key software priorities. MolSSI will be sustained by a mix of labor contributed by the community, revenue from education programs and license revenues. In summary, the MolSSI's ultimate impact will be in the translation of basic science into future technological advances essential to the economy, environment, and human health.

This project will investigate how coastal erosion in low elevation environments is linked to seasonal precipitation patterns, how humans perceive their vulnerability to erosion risk, and how humans secure livelihoods in the face of erosion. Rural populations in coastal zones are dependent on land resources to support agriculture and on marine resources to support fishery-based activities. Coastal erosion has potential to threaten economic and geopolitical stabilities due to the permanent nature of lost land resources that deprives population of land-based resources and access to marine resources. Variation in the location and timing of erosion events produces uncertainty for populations who must negotiate erosion threats without the benefit of consistent science-informed information regarding risk and vulnerability. This project contributes to a better understanding of the intersection of coastal environmental change, lowland coastal populations, atmospheric science, and economic development. Results will inform coastal processes, as well as human vulnerability and resilience, and will also make methodological advances in quantitative geospatial analysis. Students will be trained in STEM science and project elements will be used to enhance modules of existing courses at two universities. Project results will be disseminated in journals and other outlets, presented to government agencies and NGOs, and shared with local stakeholders. The project will become registered with the Resilience Connections Network, an international virtual space for the interaction between global and local leaders on resilience science and practitioners.

The study site for this project is the delta of a large river that is the outlet for water discharge driven by monsoon precipitation in a large watershed with a shoreline particularly vulnerable to erosion. The region has high rates of annual erosion whereby shorelines may retreat over 100 meters per year causing livelihood disruption and displacement of farming and fishing households. The project has three objectives. The first objective characterizes time-space patterns of shoreline erosion and monsoon dynamics at multiple scales. Understanding erosion rates at annual scales is particularly important to characterize vulnerability because households need to forecast and respond over relatively short time horizons that are not resolved at coarser decadal scales. The second objective assesses predictive relationships between monsoon precipitation and erosion dynamics with the hypothesis that time-space patterns of rainfall can help predict erosion behavior. A third objective assesses social vulnerability, risk perception and how resource endowments and adaptive behavior may promote resilience. These objectives will be addressed using a mixed-methods approach that combines geospatial analysis using earth observation data, statistical and machine-learning predictive modelling, and quantitative and qualitative primary social data collected in two villages. Investigators will produce an approximately 50-year time series record of shoreline change that will used in concert with monthly precipitation data to develop an annual predictive model of erosion risk. Field research will be done to collect social data from local populations in terms of their risk perception, vulnerability, and adaption behaviors by conducting a household survey, focus groups, and key informant interviews. Although the project will focus on the delta region of the Ganges-Brahmaputra-Meghna river system in Bangladesh, the research will provide new insights for dealing with coastal erosion in densely populated coastal zones in other regions, including the United States.

In response to the growing COVID-19 pandemic, the Molecular Sciences Software Institute (MolSSI) will leverage its position as a neutral commodity resource to help the global computational molecular sciences community quickly provide their scientific data and expertise to address the COVID-19 crisis. The MolSSI is jointly supported by the Office of Advanced Cyberinfrastructure and the Divisions of Chemistry and Materials Research. The centerpieces of this engagement will be (1) a centralized repository for simulation-related data targeting the virus and host proteins and potential pharmaceuticals, and (2) a select set of MolSSI Software Seed Fellowships for Ph.D. students and postdocs targeting COVID-19 related software tools that operate on the data developed in the repository. These two components will enable the biomolecular simulation community to share and utilize key data and other resources to help identify the structural and dynamic characteristics of the host-virus complex to generate potential leads for therapeutics. Although this project is intended to address the acute COVID-19 crisis, in the near term, it also will impact research communities and the next generation of computational molecular scientists in the confrontation and proactive resolution of future world problems.

The MolSSI will create and curate a large-scale repository containing: simulation input files (structures, configurations, scripts, Jupyter notebooks) in an organized structure; MD trajectories, analysis tools, and ready models for drug discovery; pointers to preprint servers such as arXiv, bioRxiv, and ChemRxiv on biomolecular simulation research in regards SARS-CoV-2; and DOI services that create citable data. In addition, it will engage the molecular sciences community through a set of Software Fellowships for graduate student and postdocs to carry out software development, such as large-scale MD simulations, design of drug discovery tools such as docking, machine learning for small molecule toxicity predictions, and methods for determining whether new drugs are bioavailable or can be synthesized. Collectively, these resources will speed the identification and development of leads for antiviral drugs, analyzing structural effects of genetic variation in the SARS-CoV-2 virus, and inhibitors that can disrupt protein-protein interactions to viral entry into cells and adherence to surfaces that cause disease spread.

This award is being funded by the CARES Act supplemental funds allocated to CISE and MPS.

This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.