In this issue of Scientific American, the cover article, "Cybernetic Cells," discusses how computer modeling is an absolute must when it comes to deciphering all the data from the decoded human genome. W. Wayne Gibbs reports that cells are so complex even a supercomputer model might not deliver perfect results - but it will still lead to breathroughs. Machines begetting machines? Well, birds do it and bees do it. Moshe Sipper and James Reggia report in "Go Forth and Replicate" that the latest computer simulations suggest machines could do it too. Scientists have found a cheaper way to solve tremendously difficult computational problems: connect ordinary PCs so that they can work together. In the second part of the series on Next Generation Supercomputers, William Hargrove, Forest Hoffman and Thomas Sterling explain how a low-tech approach results in "The Do-It-Yourself Supercomputer." The scientific method involves making hypotheses, testing them and getting peer review. Rarely does it result in ostracism. Virologist Peter Deusberg has doubts about the link between HIV and AIDS. Now, reports W. Wayne Gibbs, Duesberg is taking on cancer. Who says venture capitalism is dead? Certainly the CIA doesn't think so. Daniel Dupont reports in "The Company's Company."
©2001 Scientific American