The harsh way to review an article:
A man named Jinwu Zang (China Drugs Head Fired Over Article Row NATURE vol. 498 no. 7454 June 20, 2013 page 283) has got into trouble. He published a treatment on multiple sclerosis, his specialty. There is a protein, the interleukin-7 receptor that has been linked to the disease and the paper purported to show that a variant of the protein is indeed the cause.
There was a problem with a caption. As you must know, when I write I include any illustration at the appropriate place rather than say “fig.1” and referring elsewhere. I really have no need for captions at all and I find the usual way of doing it awkward. Similarly if I have a reference it is right there in the text in parentheses so it can be skipped or the actual title and date can be seen right away. Squirreling the references away in a block at the end, usually in a tiny font, serves no good purpose. The erroneous caption did not affect the results of the work. It was an honest unimportant mistake.
Another caption figure was duplicated instead of a different figure being used. Same defense. Honest mistake. Poor way of managing things anyway. If everyone did it my way the rate of such things happening would be lower. Another lab was unable to duplicate the result but there was a difference in the exact protocol: give me a break. The field of making antibodies against specific proteins is hag ridden with results that don’t get duplicated even with an identical protocol. As one worker has said, “You are at the mercy of the rabbits” (which are used to make the antibodies). Zang says the whole witch hunt (my own impression) is bizarre.
Wow. Isn’t science all about holding up results so others can try repeating them and maybe improving on them? Aren’t mistakes the common condition of mortal kind?
Some days I lament that I have no academic standing from which to draw attention to my work. But at least nobody is going to throw me out of my lab no matter how stupid and careless I become. I am not a fighter. I guess you need to be.
There have been 71 visitors over the past month.