Department of Computer Science
Brandeis University
Waltham, MA 02254
Phone: (617)736-2702
Fax: (617)736-2741
Email: jc@cs.brandeis.edu
URL: http://www.cs.brandeis.edu/dept/faculty/cohen/
Jacques Cohen is the Zayre/Feldberg professor in the Department of Computer Science at Brandeis. He holds doctorates from the University of Illinois, in Engineering, and from the University of Grenoble, France, in Computer Science. Professor Cohen has been a member of the Brandeis faculty since 1968 and he has chaired the department for the past twelve years. Last year Professor Cohen was appointed Editor-in-Chief of the Communications of the ACM .
Dr. Cohen's research interests are in languages, especially those used in AI, compilers, computer-assisted analysis of programs, and parallelism. In a 1967 paper, he proposed one of the first approaches for using virtual memory in Lisp. In 1973, he developed one of the first compiler-compilers. Professor Cohen is currently doing research on parallel compilers for constraint languages.
Dr. Cohen has published over sixty papers in refereed journals. About twenty of these publications were co-authored with undergraduates. He is the author of well-known surveys on memory reclamation, non-deterministic algorithms and logic programming. He helped organize one of the first tutorials on Prolog in the US and the first US workshop on constraints.
I will describe several tools enabling their user to estimate the efficiency of Pascal or C-like programs (including parallel ones). The approach consists of generating symbolic formulas expressing the efficiency of the programs being analyzed. The formulas are applicable to a variety of compiler-machine configurations and can be manipulated using symbolic packages like Maple. The proposed approach reduces considerably the amount of benchmarking needed to analyze programs. Several examples will be presented showing the applicability of the tools.
I will demonstrate the advantages of logic variables and constraints by:
In this talk I will present various aspects of the relationship between Computer Science and Bio-Informatics. The latter is a recent interdisciplinary field developed from the need of biologists to store and easily retrieve huge amounts of information gathered from experiments. The genomes of various organisms including humans are now available in databases. In contrast, very little information is available about the functions (behavior) of genes. The DNA information thus represents the tip of the iceberg of what needs to be done to understand how a cell works, how diseases develop, and how drugs can be designed to counteract those diseases. It is likely that the next centuries will be devoted to those tasks.
Computer science should play a crucial role in the development of bio-informatics. I will review the present objectives of biologists and describe the areas in Computer Science that are likely to contribute significantly in solving some of the biological problems.
The ultimate goal of bio-informatics is - given the DNA of an organism - simulate a cell of that organism in a computer, thus replacing in-vitro and in-vivo experiments by in-silico ones. We are far from that goal but that is the general direction that will be pursued.
In this talk I will describe how one can attempt to infer the function of genes from existing DNA and protein data. Those functions are the first step towards the simulation of a cell.
No knowledge of biology is required, but biologists are welcome.
The talks are given in English, French, or Portuguese. Each talk requires an overhead transparency projector. Copies of the overhead slides can be provided for distribution.