“But I really hankered after getting back to academia" - Why Phil Bourne chose UVA

 

Traveling the Road from 3D Protein Structure to Computation

 

Shape plays a remarkable role in human biology.

A striking thing about DNA, for instance, is its highly efficient configuration. Stretched out, the 3 billion base pairs in a strand of human DNA extend 6 feet.  Tightly wound into a double helix, the human genome fits into a space just six microns across.

Shape also plays an essential role in drug discovery. The 20,000 to 30,000 unique proteins produced by the human genome—responsible for the structure, function and regulation of the body’s tissues and organs—fold into approximately 1,200 different shapes.

Flaws in sequence or shape can cause disease. It takes intense computation to discover the small molecules that can correct or compensate for these flaws. Philip Bourne, the newly appointed Stephenson Chair of Data Science and director of the University’s Data Science Institute (DSI), is drawn to this challenge.  

Throughout his career, Bourne has been at the forefront of those creating a computational basis for understanding proteins. He is the co-developer of the Combinatorial Extension (CE) algorithm for the three-dimensional alignment of protein structures. He was co-director for the RCSB Protein Data Bank, a fundamental resource for drug discovery and medical research.

“I’ve been fortunate in that computation for me was a natural outgrowth of my interest in proteins,” he says.

In addition to his leadership of the DSI, Bourne joins the Department of Biomedical Engineering as a full professor. He was formerly the associate director for data science at the National Institutes of Health.

******

The interplay of the sequence of amino acids that make up a protein and its convoluted shape is one of the chief reasons that drug discovery is so difficult. These two factors determine the characteristics of binding sites, regions on a protein at which specific molecules or ions may interact, changing the activity of the protein.

The goal of computational pharmacologists like Bourne is to identify a drug that targets a specific binding site associated with a disease.

While this is extremely difficult in itself, another problem is that the combination of structure and folding in a series of different proteins with different functions might yield similar binding sites. The result: a candidate drug that effectively treats a disease cannot be used because it causes serious side effects.

“There is a flip side to this situation,” Bourne notes. “The existence of alternative binding sites means that you may be able to repurpose the drug to treat a different condition.”

Over time, Bourne and colleagues have developed a process using techniques from structural bioinformatics and chemical informatics that has enabled him to explain side effects of existing drugs, propose alternative uses for them and help select among lead compounds for clinical trials.

Among his achievements are insights about the side effects caused by select estrogen receptor modulators, a class of drugs that includes the anti-cancer treatment tamoxifen and a proposal that two drugs used in treatment of Parkinson’s disease could be effective against extreme drug resistant tuberculosis.

*******

Given this background, Bourne feels that the biomedical engineering department will be an excellent home for his work. “We do systems pharmacology,” he says. “We are trying to understand the implications of what we find relative to the molecular structure of the whole living system. The systems biology work that other members of the department are doing — their work on signaling pathways, protein-protein interactions and networks — is the perfect complement to what we do.”

Bourne also has found uses for his techniques that are far afield from drug discovery, thanks to the inspiration of graduate students interested in applying their methods to evolution. Through analysis of a protein’s three-dimensional structure, they were able to determine how its folds and its binding sites have evolved over time. “We think these changes correlate with alterations in the geochemistry of the Earth,” Bourne says. “As geochemistry changes, the way proteins bind metals and even the metals they bind to changes. It’s confirmation that the Earth’s evolution and our evolution are inextricably linked.”