John Matulis is a senior scientific analyst/programmer for SmithKline Beecham Pharmaceuticals (SB) at their main U.S. research and development site in Swedeland, Pennsylvania, in suburban Philadelphia. His group, Scientific Computing Consulting (SCC), part of Scientific and Clinical Systems, is made up primarily of three people: John, whose background is in mathematics, a physical chemist and a mathematical biologist. Occasionally the group includes contractors and one or two co-op students.
"SCC applies mathematics and computation to various problems in pharmaceutical research," says John. "Past and present tasks include modeling cancer tumor progression, assisting the structural/physical chemistry group in developing image analysis tools for MRI, and the mathematical modeling of the kinetics of a series of biochemical reactions. Currently, most of our efforts are spent in support of and collaboration with SB Bioinformatics to analyze human genome data. I have been working almost exclusively in the Bioinformatics effort, including doing much of the modeling and programming (primarily MATLAB, SQL and C, in UNIX and Macintosh environments)."
His group finds that they are applying mathematics to an ever increasing number of situations and problems. Mathematics is used not only in the development of computer models, but also to gain insights into new situations.
John has an A.B. in mathematics from the University of Pennsylvania, and an M.A. in mathematics and an M.S. in computer and information sciences from Syracuse University. He is currently enrolled at the University of Pennsylvania in the Systems Engineering department, working towards a Ph.D.
"I have been with SB for 10 years," John says. "I originally applied for and was hired into a position as a programmer/analyst in Clinical Systems Development (CSD). The position, building systems and programs to handle the data from clinical trials, did not require any mathematical training, although it demanded wide-ranging technical skills and programming proficiency. As departmental functions changed and were reorganized, CSD was to become less technical and more business-process oriented. So, I asked for an opportunity to work with SCC, and after about a year of being "on loan" to them, officially joined the group."
"Although I have taught full-time at several colleges," continues John, "and I still teach part-time at night, working in industry allows me to (a) more quickly see my work used or implemented, (b) learn more about different areas of science by actually participating in solving problems in those areas, and (c) continue learning without having to specialize in a narrow research area."
Some courses he recommends for students interested in his area are differential equations (ODEs and PDEs), probability and statistics, numerical methods, optimization. Also somewhat useful are certain computer science courses involving neural nets or discrete mathematics.
"A knowledge of chemistry or biology is helpful," he adds, "both for an understanding of how to model such systems, and for being able to talk to other researchers."
His advice for preparing for a career is to "obtain as broad an education as possible at the undergraduate level, including more than just a handful of humanities and social science courses. This is usually easier to do in a liberal arts program than in an engineering school. I concede that a little depth (at least 1 or 2 courses past the standard first-year course) in an 'applicable area' (e.g., biology, chemistry, economics/finance, linguistics) would be helpful, but choose on the basis of what you would like to learn about, not what you think your future job might require (no doubt, these will coincide in some cases). There is plenty of opportunity to specialize in graduate school and to focus on a particular industry or business area or personal need at that time. Additionally, as our co-ops will attest, internships or co-op positions are great experience. Most importantly, be eager to learn, intellectually curious and flexible -- much of what you need to know (past a certain basic level) is learned on the job anyway."
Return to Archived Profiles and Forums
