I recently came across an article in the Journal of Cell Biology (JCB) entitled "Quantitative proteomics identifies a Dab2/integrin module regulating cell migration" (JCB 2009; 186:99–111). Finding images with colocalization in it, I expected them to be analyzed quantitatively as well. That was a wishful thinking. On Figure 5 of the article, Teckchandani et al. counted the number of colocalized and non-colocalized particles of Dab2 an integrinbeta1 and made conclusions about their ratio in ventral versus dorsal surfaces. How they determined coloalized particles? Visually. Non-colocalized ones? Also visually. How did they compare them? Visually as well! What a visual (and totally erroneous) approach! Ironically, the paper uses word "quantitative" in the title. But that is not for quantifying colocalization, that's for other things. Why? This is a good question.Replying to my inquiry, Dr. Cooper of Fred Hutchinson Cancer Research Center shared a truly contemporary and very scientific attiude by saying that he believes only in what he sees and never relies on a "clever computer stuff". Is this disbelief applicable to the mentioned above paper he co-authored as well?
It is known that JCB not only sets trends in cell and molecular biological research, but also devotes a meaningful portion of its pages to the articles describing how to interpret and analyze biological imagery. It also explains the benefits of quantification of data and encourages readers to use it in their research. Worth mentioning in this regard are "Seeing is believing?" (JCB 2006; 172:9-18) and a very recent "Accuracy and precision in quantitative fluorescence microscopy" (JCB 2009; 185:1135-1148). So why JCB published a paper that ignores its own policies? This is an another good question.
Conclusion? Some people who publish their efforts in JCB should probably read the journal as well ... or choose a less visible venue for their misguiding publications.
No comments:
Post a Comment