RRID use fixes one easy cell line problem
- 09:26am January 29, 2019
- Anita Bandrowski
We are sure that you have seen the press stating that many of the cell lines used by scientists are actually bad. They can be misidentified (someone at some point put the wrong label on a vial), contaminated (someone sneezed), or over grown by other cells (in the immortal words of Oprah in the immortal life of Henrietta Lacks,
'HeLa's coming after you').
Researchers who run the
RRID initiative, the
ICLAC consortium, and the
Cellosaurus database decided to find out if they could radically change how many 'bad' cell lines were used in papers. So they did some cool things.
First they tried to figure out how many cell lines that researchers used in total. Through some fancy text mining algorithms they found their number.
Then they asked, ok how many of these are bad, and of course how many papers does this affect? The answer is entirely too many, 16% of papers, by their estimate, use at least one cell line that is on the bad list.
So how do you fix the problem?
Their answer was: Put a Giant, Red, Flashing warning (ok just red), where the authors are looking before they publish their paper. For this they had to ask the nice people who run journals to pitch in.
Out of all the papers published that had RRIDs marking cell lines, turns out that only about 6% were on the naughty list. Ok so you might say "why not 0?" and the answer is that probably being on the naughty list does not mean you are actually naughty. This is also true for the rest of the papers, but a drop of 60% in the use of potentially bad cell lines is something that anyone interested in reproducibility and scientific progress should be interested in.
Some places where this story was picked up: