Forgot your password?
typodupeerror
Biotech Science

Stem Cell Research Paper Recalled 112

Posted by CowboyNeal
from the making-the-grade dept.
MattSparkes writes "One of the best-known stem cell papers describes adult cells that seemed to hold the same promise as embryonic stem cells. Now some of the data contained within the paper is being questioned, after staff at a consumer science magazine noticed errors. It shows how even peer-reviewed papers can sometimes 'slip through the net' and get to publication with inaccurate data."
This discussion has been archived. No new comments can be posted.

Stem Cell Research Paper Recalled

Comments Filter:
  • No shit sherlock. (Score:5, Informative)

    by Watson Ladd (955755) on Thursday February 15, 2007 @08:29PM (#18032768)
    Peer review isn't about the accuracy of the data. It's about how the data supports the conclusions and whether the paper is notable enough for the journal. Peer reviewers don't have the time and money to duplicate every experiment they review.
  • by CapsaicinBoy (208973) on Thursday February 15, 2007 @08:33PM (#18032816)
    This is exactly how the system is *supposed* to work. Dr. Verfaillie publishes her team's findings, and others try to replicate it. If they can, the original finding is supported. If not, the failure to replicate usually leads to other insights. My old boss was usually more excited when an experiment failed than when it worked, because was what led to breakthroughs instead of mere confirmation.

    Yes, the process can take time, and god forbid you were the poor grad student that spend 3 years heading down a blind alley, but this episode just reaffirms that overall, the process works.

  • by bobscealy (830639) on Thursday February 15, 2007 @08:35PM (#18032832)
    RTFA. New Scientist noticed that 6 graphs were duplicated exactly across two paper that were describing different cells. They questioned the results, and a panel of experts reviewed the data. Nowhere does it say that New Scientist magazine got their hands on even a single stem cell.
  • not just "sometimes" (Score:3, Informative)

    by SuperBanana (662181) on Thursday February 15, 2007 @08:49PM (#18033028)

    It shows how even peer-reviewed papers can sometimes 'slip through the net' and get to publication with inaccurate data."

    How about the misandry-filled assertations about how women use more words than men, have brains more "wired" for communication, etc? Turns out that everyone's been quoting each other and nobody can even stick to facts interview to interview [boston.com].

    What's appalling is that the author, Dr. Brizendine, not only holds a top academic position, but also has a best selling book that is full of "facts" that are complete fabrications.

  • Re:No shit sherlock. (Score:3, Informative)

    by rumblin'rabbit (711865) on Thursday February 15, 2007 @08:51PM (#18033044) Journal
    Amen to that. I've peer reviewed papers, and for most part you end up trusting the authors. It's not like the reviewer can rerun the experiments or inspect the raw data.

    Much of peer review involves checking the form, rather than the substance, of the paper. Does the paper follow proper protocal? Is it clearly written? Are the references complete and correct? Should it be shortened or added to?

    The substance of the paper also comes into it, of course, but the reviewer is very limited as to what he or she can do when it comes to checking the validity of the claims.
  • by LordLucless (582312) on Thursday February 15, 2007 @09:30PM (#18033448)
    The whole "stem cell debate" thing makes me embarrassed to tell people that I'm American.

    See, people like you trolling a debate they know nothing about would make me ashamed to be an American (if I was one). Here's a hint: this story is talking about adult stem cells, which has no significance at all in regard to the current political/moral question of embryonic stem cells. They're two totally different things, obtained through totally different processes.
  • by CapsaicinBoy (208973) on Thursday February 15, 2007 @09:59PM (#18033714)

    people like you trolling a debate they know nothing about would make me ashamed to be an American (if I was one). Here's a hint: this story is talking about adult stem cells, which has no significance at all in regard to the current political/moral question of embryonic stem cells.

    Actually, they are completely relevant to the moral/political question of embryonic stem cells, in so far as embryonic stem cell opponents have been using these adult stem cells to have their cake and eat it too.

    Specifically, they've been tying the hands of researchers due to their religious beliefs and then shielding themselves from criticism by claiming "oh, we don't need embryonic cells anyway, because adult stem cells are just as good." Here is one example [nationalreview.com].

  • by EmbeddedJanitor (597831) on Thursday February 15, 2007 @10:08PM (#18033784)
    For a very illuminating read on this whole business, have a look at "Fabulous Science" by John Waller. Even illustious organisations such as the Royal Society - set up for goal of providing peer review - have been conned or biased (political correctness, partiotism, religious beleifs...). Even Nobel prizes have been awarded for massaged experiments.
  • Re:No shit sherlock. (Score:3, Informative)

    by imkonen (580619) on Friday February 16, 2007 @10:00AM (#18037568)
    "Peer review isn't about the accuracy of the data. It's about how the data supports the conclusions and whether the paper is notable enough for the journal."

    You beat me to it. A lot of journals only use 1 or 2 reviewers. 3 is the most I've ever heard of. Reviewers are neither paid for their work nor given a lot of time to do it, so some really just phone it in. Even the more thorough ones: we're talking essentially 3 chances to find a mistake. I've rievewed papers and if I notice a mistake I'll certainly point it out even if it's just a grammar, but I'm not going to lose any sleep over the fact that I didn't manually overlay their graph with every other paper they've written to check for a mistake like the one mentioned in the article. That's the kind of mistake it takes 10000 pairs of eyes to catch.

    To put it in terms a typical slashdotter should understand, that would be like expecting a program to be completely bug free after 4 revisions. Of course part of the process is that after it's been published, it's expected that other scientists will look at the paper and catch things reviewers might have missed...kind of like the open source coding model.

The only thing cheaper than hardware is talk.

Working...