XSEDE EMPOWER Positions

Please note that XSEDE EMPOWER ended with the conclusion of the XSEDE project on August 31, 2022. The following information is provided for archival purposes only.

This page provides a listing of positions that are under review for the next iteration of the program, as well as those that have been approved for current and previous iterations of the program. As part of the review and approval process, accepted student applicants are associated with an approved position.

If you are intending to mentor a student, you should create a position. If you also have a student in mind with whom you want to work, that student should submit an application. If you intend to mentor multiple students, please create a separate position for each student.

Project Title / Status	Summary	Creator / Institution
Computational Analysis of Copper Intermediates in Reduction of Nitrous Oxide Completed	An intern will conduct quantum mechanical calculations of copper enzyme intermediates and calculate X-ray emission spectra. These calculated spectra will be compared with calculated spectra of model complexes to determine whether different coordination modes of nitrous oxide and reduced species can be distinguished by XES to identify enzyme intermediates.	S. Chantal E. Stieber California State Polytechnic University-Pomona Pomona, CA
A Visual Answering System for Crossword Puzzles using Deep Learning Completed	We are currently witnessing an excitement in the research and industrial communities regarding advances in AI. In this project, we target an AI supporting system for crossword puzzles to provide visual aids through the image captured by a mobile phone. The system will tie together skills and knowledge in deep learning and computer vision. Our past efforts are investigating the approaches of deep learning and computer vision for crossword puzzle grid detection and alphabet letter identification. Our following research will continue to focus on developing deep learning models to solve crossword puzzle problems. We would like to apply for an XSEDE empower position, which will allow us to continue studying, developing, and training effective deep learning models for this AI-driven crossword puzzle system.	Hao Ji California State Polytechnic University-Pomona Pomona, CA
Structural Prediction Using Machine Learning Completed	While it is fundamentally accepted that structure affects function, predicting subsequent changes in function from perturbations to the structure poses a challenge. In a structure as complex as a protein, mutations in the molecule's structure often result in diminished or even no protein activity. The protein p53, for example, is responsible for the destruction of its own cell upon the detection of heavily damaged DNA and plays a critical role in the suppression of tumors. However, mutations in the protein have shown decreased activity and thus adversely affects the body'™s capability to resist tumor growth. Interestingly, some second mutations, referred to as "rescue mutations", have been shown to partially restore function to these mutated proteins and by extension their ability to suppress the spread of cancerous cells. By predicting which structural regions to alter in order to elicit functional change, the development of rescue mutations similar to p53 would allow for a more targeted approach to the development of treatments on a molecular basis. Molecular dynamic (MD) data has been generated for a protein, Cyclophilin A. This protein is responsible for the transition from cis to trans state of proline residues in protein backbones, playing a crucial role HIV development in hosts and immunosuppressant in organ transplants. Using the generated data for this system as well as the data from several mutated forms of this system, a supervised learning network will be developed to predict functional effects of structural changes. A scoring function will be used to extract structural information from the generated MD data, denoting intramolecular contact between different residue pairs. The data set is a SSV file representing a time series, representing contacts between two residues at certain specific frames in the simulation (Fig 1). As the data has been reduced to a binary state, information is given by the contact dynamics in the system. As residues contributing to primary and secondary protein structures will be in contact for the entire simulation, they will convey little information about the system's dynamics; conversely, distant residues will have no contact and contribute very little to the system's dynamical structure. Thus, only residue pairs with contacts occurring more than 10\% and less than 90\% of the time will be considered, referred to as dynamic contacts. By learning the dynamic structure of a controlled system, the subsequent structure of an altered system can be predicted and serves as a means to intelligently target mutations. Using the Keras framework in Python, a recurrent neural network will be used to predict the contacts state of the nth frame from previous data. As Keras is a highlevel framework, rapid prototyping allows for the quick testing and implementation of learning networks. The choice of a recurrent neural network is due to its ability to retain past information as context. More specifically, a LSTM network will be invoked due to its ability to better handle long-term dependencies in the data (Fig 2). Calculations will be conducted using the allotted nodes of Dr. Edirisinghe on the Bridges, Comet, and XStream systems.	Neranjan Edirisinghe Georgia State University Atlanta, GA
Undergraduate Researcher Assistant for High-throughput DFT Study of Functionalized Polycyclic Aromatic Hydrocarbons (PAHs) Completed	The persons hired in this position will be expected to perform high-throughput electronic structure calculations (in the DFT format) in order to understand the effect functionalization has on the electronic properties of PAHs.	Bohdan Schatschneider California State Polytechnic University-Pomona Pomona, CA
Simulation of Biological Neural Networks Completed	Write parallel simulations of biological neural networks. Work with standard processors and GPUs. Evaluate computational intensity and efficiency.	Chris Fietkiewicz Case Western Reserve University Cleveland, OH
Machine Learning Over Parallel Architecture and Outreach Completed	We hope to continue our wonderful EMPOWER student intern from Fall 2018 semester, and ultimately guide him towards a career pathway in computational research.	Xinlian Liu Hood College Frederick, MD
Visualization of Turbulent Events in Wall-bounded Flows via Virtual Reality Completed	Turbulence is mainly characterized by randomness, chaos and disparate range of turbulent scales and mixing. Despite its chaotic behavior, investigation performed during the last six decades has conclusively demonstrated the presence of organized motions in turbulent boundary layers, so called coherent structures (CS). These structures can be considered the building-blocks of turbulent boundary layers, and significant attention has recently been given to explanation of their creation, development and destruction. This project involves time-dependent 3D scientific visualization of coherent structures and turbulent events from a Direct Numerical Simulation (DNS) database of spatially-developing turbulent boundary layers subject to streamwise favorable pressure gradient by means of a fully immersive approach or virtual reality (VR). The main purpose of the EMPOWER internship project will be to develop a post-processing code in C++, Blender and Unity platforms with GPU capabilities in order to perform flow visualization in a virtual reality environment.	Guillermo Araya University of Puerto Rico-Mayaguez Mayaguez, PR