Something I find annoying is when people publish without citing earlier sources.

For instance, in the mid-1990s my then boss asked me to write code to do what he called "annulus" measurements. The idea was to measure the intensity at the perimeter of an object (cell stained for specific proteins) and then to step in at one pixel intervals and measure the intensity as a function of distance from the perimeter. Because this was for fluorescent microscope images, the software also looked outside the object for background subtraction. While we could get okay tracing of the objects by thresholding or looking for contrast gradients, we found that hand tracing, even by a wide range of people, was more precise.

We wrote the code as a macro in NIH Image (and could not have done so without extensive help from both the NIH-Image listserv and Wayne Rasband) and, as far as we know, were the first to publish this method in Chan 1998. This was subsequently used as a standard tool in other published manuscripts, rewritten in ImageJ, and distributed to other labs by the Analytical Imaging Facility at the Albert Einstein College of Medicine. Functions that we struggled to code as functions within the macro language were incorporated as standard functions into ImageJ. This was great.

But then in early 2015 we came across a paper Barry 2015 that explained this same method for analyzing the same proteins as our initial application almost two decades ago and didn't even cite us. At some point methods may become so common that people stop quoting earlier sources, but given that this is both not such a common method and that the biology is so similar, we were surprised to not even get a mention. Also, the new published method touts the benefits of automatic segmentation without noting that as sweet at automation may seem, sometimes it just isn't as robust as human interaction.

Has anyone confirmed that the new published method works? Obviously, we don't know who the reviewers were. Does this journal, and do other journals, require testing the software as part of the review process?

movie generates

J Cell Biol. 1999 Apr 19;145(2):331-45.
Relationship between Arp2/3 complex and the barbed ends of actin filaments at the leading edge of carcinoma cells after epidermal growth factor stimulation.
Bailly M1, Macaluso F, Cammer M, Chan A, Segall JE, Condeelis JS.

EGF stimulates an increase in actin nucleation and filament number at the leading edge of the lamellipod in mammary adenocarcinoma cells.
Chan AY, Raft S, Bailly M, Wyckoff JB, Segall JE, Condeelis JS.
J Cell Sci. 1998 Jan;111 ( Pt 2):199-211.



last updated 20150803_1715