George Casella's work within the statistics department has led to interdisciplinary projects and studies across the University of Florida.

Above: George Casella's work within the statistics department has led to interdisciplinary projects and studies across the University of Florida.

Share This Story

Bookmark and ShareBookmark This
EmailEmail Story

Follow CLAS

Subscribe to the CLAS RSS Feed RSS Feed
Become a Fan of the CLAS Facebook Page FACEBOOK
Follow us on Twitter TWITTER

George Casella

Head of the CLAS

Identifying genes that cause disease in the pine trees, devising algorithms for radar detection of land mines — it’s all in a day’s work for George Casella, Ph.D., distinguished professor of statistics with the College of Liberal Arts and Sciences and the University of Florida Genetics Institute.

“I like to work on as many different things as I can,” Casella says from his office at Griffin-Floyd Hall.

Fresh from a visit to Spain, Casella explains how he uses a technique called the Monte Carlo Method to help solve a vast array of problems. Scientists at Los Alamos National Laboratories used the method during the race to build the atom bomb during World War II, Casella says.

Its power is that it allows experimenters to solve the problem they are interested in, not just problems that have a nice mathematical solution, he says.

“Instead of getting an answer by doing a very difficult mathematical calculation, you devise a simulation that mimics the equation,” he says. “By doing that simulation thousands or possibly millions of times, the average is going to be your answer. We can take advantage of today’s bigger and better computers and take a bypass to get to the end of these enormously complex calculations.”

Monte Carlo techniques can be applied across the scientific landscape, which has led Casella along some interesting paths.

The list goes on. Does living near a power plant increase your cancer risk? Can we genetically engineer better, more nutritious food? The questions are limitless, but statistics can be applied to all of them.

“Generally speaking, statistics is the science of drawing conclusions from data, and it doesn’t matter what kind of data it is,” Casella says. “It is not to be confused with bioinformatics, which is more akin to computer science and data management — pulling together information about plants, animals, people, DNA sequences and so on. Statistics tells scientists how to collect data and set up experiments so that in the end, they can draw a legitimate conclusion about a discovery that can hold up to scrutiny from peers.”

His focus on developing statistical algorithms that are designed to answer questions in genetic analysis make him a major contributor to the bioinformatics effort of the UF Genetic Institute, says UFGI Director Kenneth Berns, M.D.

When Casella arrived at UF in 2000, he quickly found likeminded colleagues in genetics and genomics at IFAS and elsewhere, which formed the beginnings of a functional genomics discussion group.

“We decided we were going to meet once a week and read papers about subjects that caught our attention,” Casella says. “Whoever wanted to be there could come. If someone were interested in gene-environment interactions, we would say, OK, let’s read some biology papers, some statistics papers, maybe even analyze some data. We’ve essentially been doing this for ten years now, and that’s how I met lots of people across campus who work with plant, animal and human genetics.”

Casella’s accomplishments in genetics and genomics were noted by the Spanish Royal Academy of Sciences, which last year bestowed him with Spain’s highest scientific honor — election into the Royal Academy of Mathematics, Physics and Natural Sciences.

Now, as a corresponding foreign member of the academy, Casella joins distinguished scientists who have won Nobel prizes in physics and chemistry, as well as probability theorists, statisticians and experts in operational research such as James O. Berger, director of the Statistical and Applied Mathematical Sciences Institute at Duke University; Thomas L. Saaty, a mathematician at the University of Pittsburgh; and Simon French, a mathematician at the University of Manchester.

The academy cited Casella’s relationship with Spanish scientists through joint work on projects and scientific publications, including collaborations with professors Javier Giron of the University of Malaga, and El?as Moreno of the University of Granada, who are also academy members.

His collaborations with scientists in Spain go back to 1987, when Casella was invited to a conference at the University of Valencia by Jose Bernardo, a practitioner of Bayesian statistics.

Bayesian inference – a way to determine the probability of a scientific hypothesis or assumption — fits hand in glove with Monte Carlo methods in that it is a way to arrive at a complex answer by using layers of more simple calculations.

Besides that, Casella says, “I was really interested in the Spanish culture and people.”

With little command of the language, Casella set out to learn enough Spanish to communicate. One conference led to the next, and Casella became a close associate with Moreno, another statistics practitioner.

Casella wound up teaching more courses, returning to Spain year after year. In the meantime, he began publishing papers with Giron and Moreno— “great statisticians who are devoted to their work,” Casella says.

By 2002, Casella moved his family to Spain for about five months. While he taught at the University of Grenada, Casella’s children were enrolled in the public schools.

“I wanted to learn more about Spain and to expose my kids, who were 8 and 10, to a foreign country,” he says. “We got there in August and I worked hard on the Spanish language to be able to teach a course in October. Giving a talk about genetics in Spanish was a stretch, especially since all I had to fall back on was college Spanish. Part of the reason I was able to do it is the Spanish people are so nice, when you start tying to speak, they just help you. When you get immersed in a culture like that, it’s easy.”

Casella currently serves as the Col. Allan R. and Margaret G. Crow term professor and is a member of the governing board of the Statistical and Mathematical Sciences Institute, a trustee of the National Institute of Statistical Science and an advisor for the Oslo Centre of Statistics for Innovation in Norway.

He received his master’s and doctoral degrees in statistics from Purdue University and he arrived at UF from Cornell University in 2000, and chaired the Department of Statistics from 2000 to 2006.

Credits

Writer

John Pastor, Health Science Center News and Communications, 352-273-5815

Photo

Jane Dominguez, Communications and Outreach, janed@ufl.edu, 352-846-2032

back to Head of the CLAS Index >>

top >>

CLAS Navigation

News, Calendar of Events, Head of the CLAS, Submit News/Event, Media

Search


CLAS Portals

Alumni
Faculty/Staff
Parents
Students

College of Liberal Arts and Sciences

2014 Turlington Hall
P.O Box 117300
Gainesville FL 32611
P: 352.392.0780
F: 352.392.3584