NMSU branding

New Mexico State University

New Mexico State University

News Center

NMSU professors working in growing bioinformatics field

Joe Song and Jing He, assistant professors of computer science, are in the forefront of the developing bioinformatics field. Song was brought to NMSU for its Center for Research Excellence in Bioinformatics and Computational Biology (CREST) and Professor He is one of the principal investigators for the CREST grant from the National Science Foundation.

Song's current research involves statistical computing, computational systems biology, neuronal signal analysis and computer vision.

"I develop efficient statistical modeling and learning algorithms to compute effectively a mathematical representation of the underlying mechanism that generates the observed data," Song said.

Among his current projects, Song is collaborating with Graciela Unguez, of NMSU's biology department, to investigate how the muscle cells of electric fish convert to electric organs (part of the nervous system) when stimulated by high-frequency electrical signals.

To determine which genes are involved in the conversion, the biology department measures levels of gene expression over time then Song uses computation to look for consistency and patterns to eliminate unimportant genes and find interactions among involved genes.

"The electric fish is a perfect model organism to study nervous system regeneration, which will allow us to gain insights on molecular mechanisms that might be applied to human nervous systems," Song said.

In another of Song's research projects, he is attempting to uncover the cellular processes which cause cancer by studying the cells of the wing of a fruit fly. Normally, cells have a mechanism that causes them to multiply for a certain amount of time and then stop. Cancer occurs when a particular cell does not stop proliferating. The cells of fruit fly wings are thought to be some of the most uniform and therefore good candidates for study.

In order to learn about the mechanism at work in a cancerous cell, Song is searching for patterns in the process of proliferation in healthy cells. This research is supported by the National Institutes of Health and is done in collaboration with Bruce Edgar of the Fred Hutchinson Cancer Research Center in Seattle, Wash.

As another member of the NMSU faculty working with CREST, Professor He's research is in the area of computational structural bioinformatics.

Researchers have long sought an experimental method to determine the structure of large protein complexes with tens to hundreds of proteins because proteins perform their activities through interaction.

"A snapshot of such a large complex can provide extremely useful information in understanding the mechanism of the proteins," He said.

Electron cryo-microscopy, which uses an electron microscope to generate density data of proteins, is a rapidly growing technique that can potentially solve this problem. However, it is clear that both experimental work and computational work are needed to achieve the desired results.

"Currently, there are more and more such data collected using this technique, but there is not a general method to figure out the protein structure hidden in these data," He said.

Without knowing the structures, the biological interpretation is limited.

"My goal is to develop a computational method to derive the 3-dimensional protein structure from a very rough description of the protein density," He said. "Structure is very much related to function. The composition and order of amino acids determine what kind of protein it is."

To begin, He searches for characteristic regions in the density map and uses them to assist the prediction of the chain of amino acids that serves as the "backbone" for the protein. The critical part of the work is to find a conformation for the protein that is both energetically comfortable and consistent with the density of the protein.

Once the chain is found, researchers can locate the "functional pocket," the most important part of the protein, which is related to a cluster of a particular amino acid.

In disease proteins, the functional pocket could be identified so a "patch" (a drug) can be placed over it, blocking its function. An additional challenge is that any patch used to block a negative protein function must not be toxic to other elements in the body.

Drug companies have shown a great deal of interest in this "rational" drug design for creating new pharmaceuticals. However, He noted that so far, no drug has been created in this manner. Drugs are currently designed by testing and screening thousands of compounds.

Although researchers may still be far from engineering drugs using the structure of proteins, progress in the field has made He hopeful.

"When it gets closer to the real world, I get more excited," He said.

Song and He believe CREST has been a great asset to their research.

"The Center has provided new incentives for interdisciplinary research in bioinformatics, otherwise impossible," Song said. "The National Science Foundation -Department of Energy Faculty-Student Teams (FaST) program has funded me and four undergraduate students to work with life scientists at Lawrence Berkeley National Lab in the summer of 2007 and Oak Ridge National Lab in the summer of 2006. The Center has also been supporting graduate research assistants to work with me. Such a substantial amount of support has definitely enhanced my research productivity."

"The Center has been quite helpful in supporting the research and education activities," He said. "The funding for my subproject has been used to support five undergraduate research assistants and 10 graduate research assistants. It was also used to support faculty and students to attend conferences."