Brave new RNA world.

In 1995, you could still find, in the opening paragraphs of manuscripts and grant applications from my lab and others, the phrase “In the absence of a crystal structure for the ribosome....,” by way of rationalizing any number of biochemical studies aimed at figuring out the three-dimensional locations of things inside the ribosome, and ultimately its mechanism. We no longer have this excuse. Today, the Protein [sic] Data Bank, or PDB, lists several dozen high-resolution crystal structures of the ribosome. Needless to say, this has dramatically transformed the ribosome field. When the first ribosome crystal structures were announced at the meeting in Helsingør, Denmark, in June, 1999, I anticipated an occasion of great celebration; after all, we had waited some 40 years for this magical moment. Instead, the atmosphere was like that of a funeral, whose participants were paying their somber respects to the deaths of their careers. There was a widespread impression that all of the main questions concerning the mechanism of protein synthesis had been answered, and we would now put the ribosome to rest. Instead, we find that the ribosome continues to challenge and humble us. Although it is true that we could not expect to understand its mechanism at the molecular level in the absence of a crystal structure, the structure itself has raised more questions than it has answered. For example, confusion and controversy continue to haunt the study of aminoacyl-tRNA selection and translocation, two of the most fundamental steps of protein synthesis. But investigators of very different political and religious persuasions are at last able to agree that the mechanism of action of the ribosome is based fundamentally on its RNA, rather than its protein components. In the 20 years since the founding of this journal, the practice of academic science itself has undergone profound changes, not all of which can be scored as improvements. Most notable has been the takeover of our lives and consciousnesses by computers and related devices. In 1995, you could still see faculty members working in their laboratories, or discussing experiments at the blackboard with students and postdocs. Now, as I stalk the hallways of my department, I find my colleagues in their offices staring into screens, typing. Students and postdocs seem to spend more time logged into their laptops (further cut off from potential interaction with their bay-mates by ear-buds fed by MP3 players) than doing actual experiments. Nor has the proliferation of kits (which shield us from the distraction of actually knowing what we are doing) helped this situation. The simple presentation of one’s progress at group meeting devolves into a weeks-long adventure into the mysteries of Powerpoint, Photoshop, and font selection. Submission of a manuscript for publication has morphed from a couple of hours of Xeroxing into a mind-numbing marathon of file formatting and figure rendering, followed by endless automated responses from the journal with requests for pdf conversions, faxed legal documents, and the details of which authors handled which pipet. In spite of the Internet, e-mail, electronic submission, and electronically-generated figures and text documents, there continues to be a delay of months between submission and publication. It is sobering to remember that a paper by Watson and Crick, published in Nature on April 25, 1953, had only been submitted on April 2nd (!), long before the existence of personal computers, FedEx, or even fax machines. Note that the 23-day interval had to have included converting Odile Crick’s drawing of the DNA double helix into an engraving or lithograph that could be set and inked by the printers. On the bright side, the emergence of the crystal structure of the ribosome almost exactly paralleled the development of the technology that was essential to solve it. The first crystallographic data that we collected from our weakly-diffracting ribosome crystals required an exposure time of several minutes per frame at the synchrotron, and an additional several minutes to read out the data from the imaging plate. Nowadays, much better data are collected with exposures of a fraction of a second and readout times of a few milliseconds. In the early 1980s, the fastest graphics computers struggled to rotate an all-atom representation of a small protein, and cost (including the enormous VAX 750 “minicomputer” and reelto-reel tape recorder required to drive it) about $250,000 in 1980 dollars. Perversely, the subsequent rapid development of computer graphics technology was not driven by the needs