This website uses cookies to ensure site visitors get the best experience on our website. By continuing to use this site, you accept our use of cookies and Privacy Statement. To find out more, please visit Southern University's Privacy Statement.

I agree
High Performance Computing Lab

High Performance Algorithm

High Performance Computing

Welcome to the High Performance Computational Biology and Material Science Lab (HPC-BMSL)

Our lab's researches are focused on the high performance computing, bioinformatics, and nanomaterial simulation.

The objectives of our lab are:

(a) developing novel high performance computation algorithms and methods to facilitate atomic level molecular dynamic simulations; predict secondary and ternary structure of proteins, protein docking, to understand life process and assist drug design; understanding and predicting the electronic, optical, magnetic, and structural properties of the selected novel electronic materials;

(b) to provide an infra-structured platform for systematically mentoring and training of under-graduate, graduate students, and post-doctors at Southern University and A & M College;

(c) to attract talented graduate faculties to SU and promoting and enhancing the interdisciplinary collaborating among SU campus(Computer Science, Education, Mathematics, Physics, Chemistry, Biology, EE and ME, CEES center) and with LBRN, LaSPACE, LONI supported six research universities(LSU ME, Louisiana Tech. material science, and other four campuses), and industries(pharmacy, NASA, and green energy related chemical engineering).

The four sub-projects will synergistically address complementary tasks to dramatically enhancing our fundamental knowledge and practical applications in the biophysics, biochemistry, drug design, and nano-size material science. The titles of the research subprojects are:

(1). A novel reduced coordinate space method for molecular dynamic simulations , by R. E. V. Finley and S. Yang;

(2). Developing algorithms for predicting the Secondary and tertiary structure of

proteins and modeling protein docking and interaction, by S. Bai, E. S. Khosravi, and S. Yang;

(3). Design and simulating h igh reflective IR and visible band TBC for NASA, a LSU(ME)/SU(CS and ME) joint research, by S. Yang, E. S. Khosravi; and

(4) Research outreach ---under-represented undergraduate and graduate student training by all the members.

The first project develops and studies a reduced simulation method (RSM) that uses a kD coordinate system defined using principal component analysis (PCA) of a standard MD trajectory to explore molecular motion. A primary objection of this sub-project is to provide a tool which allows scientists to efficiently survey molecular motion using a limited MD trajectory. The essential components of the reduced simulation method include isolating k dominant features of an MD trajectory using ARPACK and defining a kD coordinate space; constructing an approximation of a potential energy surface based on the defined coordinate space; updating coordinates and velocities in the kD space based on the approximate energy surface; and analyzing the resulting information with respect to the original 3nD coordinate space. We will apply this method to various classes of molecules to do a benchmark test. We will compare our simulations to experimental data, such as data obtained from infrared (IR) spectroscopy. The IR spectrum of a molecule shows which frequencies of IR radiation are absorbed by the molecule and can be used to identify the functional groups in a molecule. This is also information that the RSM seeks to reveal. We will compare the data by considering the Fourier transform of the velocity autocorrelation function, the power spectrum. Once we finished all the benchmark tests, we will extend and interface this method widely into bioinformatics and nanomaterial simulation utilizing readily available LONI and TeraGrid high performance computing facilities.

The second sub-project falls into research area of computational biology. Computational algorithms will be developed to predict the secondary and ternary structure of proteins. Computational modeling will be performed to simulate protein docking and interaction. In this research, two proteins, gK and UL20 of Herpes Simplex Virus Type-1, will be used as model proteins. The research is expected to achieve efficient algorithms for predicting protein structure, facilitating drug design and combat herpes virus infections. Currently we are utilizing NAMD package to simulate gk and UL20 interact with membrane.

The third sub-project is on a proposed NASA project working on design and simulating IR and visible light band high reflective TBC. Dr. Khosravi, Dr. Yang, one graduate, and one undergraduate students will work on the project.

The fourth project is research outreach: training graduate and undergraduate students, especially African-American students, at Southern University and A & M College, a traditionally large HBCU institution. The PI and Co-PIs have NSF supported STEM program and a proposed NSF outreach project. Currently there are 385 undergraduate students and 70 graduate students enrolled in Computer Science Department. They will be trained by intimately engaging them in the activities aimed at the attainment of the high performance computing technical objectives above through our carefully designed training programs. We expect many more students from ME/EE, Physics, Chemistry, and Biology Department to benefit from it by virtue of our track record in training students.

The PI, Dr. Khosravi, Chair of the Computer Science Department, is currently funded by Navy, Raytheon, NSF, BoR, NIH, and NGA. Two Co-PIs are supported by LONI and Computer Science Department. Two ongoing projects which Dr. Yang is working on are funded by LBRN and LaSPACE. Current close collaborations with CEES in SU, LSU Vet School, LSU/ME, Louisiana Tech. Material Science would generate new opportunities to attract more talented faculties and post-doctors all over the nation and the world, which without doubt fits into SU and LONI's long term development strategy.

Our proposed projects were selected by LONI recently.