Science reported the case of Elizabeth Goodwin, a former UW-Madison associate professor of genetics who pleaded guilty to a charge of scientific misconduct for falsifying data in a grant application to the NIH.
What is interesting (and sad!) though is not so much about sentence meted out to Goodwin (most likely just a fine and a ban from receipt of federal grants for 3 years), but the fate of her ex-students (ie. the whistle blowers):
The university praised the students for having done the right thing. A university investigation subsequently concluded that Goodwin had falsified data on grant applications and cast doubt on three papers, all of which were later cleared of any problems. Goodwin resigned. But the outcome for several students, who were told they had to essentially start over, was unenviable. One, Chantal Ly, had gone through 7 years of graduate school and was told that much of her work was not useable and that she had to start a new project for her Ph.D. (The reason wasn't necessarily because of falsified data but rather, Ly and the others thought, because Goodwin stuck by results that were questionable.) Along with two of the others, she quit graduate school. Allen moved to a school in Colorado. Just two students chose to stay at UW.
One of those who left reflected about the case in the Science story published in 2006. "Are we just stupid [to turn Goodwin in]?'" Sarah LaMartina said. "Sure, it's the right thing to do, but right for who? ... Who is going to benefit from this? Nobody."
The system as is right now is heavily tilted to the PI's favor. Principal investigators have too much power over the fate of their graduate students' and postdocs' scientific careers.
The PI has the power to direct the efforts of his/her students and postdocs, choose their projects, set their hours, tell them who to work with, decide whether their data is worth keeping or suitably labeled as junk, decide whether, when, and by whom their results are presented, and so forth. The PI controls all the resources the graduate students need -- funding, training, even access to other faculty in the department.
Even if he/she is to fall in disgrace, the students and postdocs won't be able to get away unscathed, as this episode demonstrates. A professor in my PhD institution once gave us (then 1st year grad students fresh out of college) this piece of advice:
"Your relationship with your PhD and postdoc advisor is most important, even more so than your spouse, especially if you stay within the scientific community. You can divorce the latter, but your link to your advisor(s) is permanent. So make sure you choose the right one."