Database- and Workflow-Management Challenges Posed by the Human Genome Project

The human genome project depends intensively on electronic collection and dissemination of information. In this respect, it arguably foreshadows much future scientific research. Its dependence on electronic collection and dissemination of information also poses strenuous challenges to current databaseand workflow-management technology. We discuss these challenges as they arise in community databases of collected genome data and in laboratory information-management systems. We survey what we see as the most important areas for research in how to better meet the databaseand workflow-management software requirements of the human genome project. 1 The Human Genome Project--Databases and Workflow Management It is reasonable to think of a living cell as a tiny biochemical factory controlled by a biochemical computer. The genome of an organism is the software for this computer. This software is encoded in a chemical called DNA, which is a linear molecule composed of basic building blocks, called bases) each of which encodes 2 bits of information. The genome contains numerous subroutines, called genes, and associated control structures, called regulatory elements, that determine when each subroutine is executed. When the hardware of the cell executes a subroutine, it causes the synthesis of a chemical product which, in most cases, is a protein. In living organisms the factory and the computer are one in the same-the software (the genome) builds the computer (the cell) that it executes on. The