"We've seen an exponential accumulation of biological data in public databases in recent years," says Prof. Jing Hu. "Because of their cost and speed, experimental approaches can hardly keep up with the accumulation of the new data. Computational approaches are widely applied to extracting information from huge amounts of data."
One computational approach is the use of an artificial neural network, a mathematical emulation of the biological neural system. Hu is using the method to predict protein-binding sites from protein sequence information.
"We want to develop an intelligence system and train it to predict the functions of other proteins," Hu says. "We want to study protein function, structure and binding sites. This is the simple story, but in practice, it is very complex. The biggest challenge is that the biological system is too complex, and current machine learning methods are still not mature enough to perfectly model the biological system."
Hu spent seven years at Utah State University, where he earned his Ph.D. in computer science. He came to the U.S. after working for three years as a telecommunications engineer at Alcatel Lucent, a French company in Shanghai. "Going to Utah was a big shock," says Hu, who grew up in Anhui Province by the Yangtze River. "Shanghai has a lot of city life and is developing fast. There weren't many people on the streets in Utah, and culturally I could not get used to it."
During his time out west, Hu took a liking to Utah's vibrant landscape. "I like outdoor activities, like fishing, hiking and mountain biking," he says as a mountain bike rests against a wall in his office in Stager Hall. "I still miss that part of Utah, and the snow."
The 2,500-mile journey to Lancaster from Utah gave Hu a chance to see the nation's natural and cultural diversity. Lancaster, he says, reminds him more of China than Utah. "I like the city and culture here," he says. "I can go to a museum, a concert, the mall or great restaurants."
On campus, Hu has enjoyed being a part of the College's developing bioinformatics program. "Students have the chance to have formal training in bioinformatics before going on to a graduate degree," he says. "Most classes are small, so I can talk to each student and get to know them. This is a great liberal arts college."