Hi Reader, what do you think about the current publishing system?
Almost every decision about an open position considers publication history, and scientists measure their status based on Impact Factor
However, more impactful papers are actually more sustainable.
Thus, let me show you how to motivate sustainable experimental design by pointing out how it can increase citations:
Today's Lesson: Sustainability By Design
How to run high impact studies sustainably
Number Of The Day
The Impact Factor measures the number of citations divided by the number of papers published within a specific journal within 3 years. Prestigious journals such as Science and Nature have IFs of 50.5 and 44.7, respectively. However, the median (most frequent) IF of scientific journals is only 2.6! Historically, the average Impact Factor was around 1.7 in 2015, 2.6 in 2020, and 3.9 in 2022.
2.6
Increasing Citations & Sustainability
In science, we trade resources to create papers that are read and provide meaningful insights for human life.
This graph is based on a preprint and reworked in a Science article. Note that all papers with ≥100 citations were subsumed under the 100 point. Side note: if you want to read more about Impact Factor distributions of journals, you can do so in this preprint.
Although it is hard to measure what a “good” publication is, the closest we can get is counting citations. So, how can you get more citations to be more sustainable and boost your career?
Descriptive Findings Are Not Enough
Less-cited papers that often just describe one observation after another and mention potential mechanisms in their discussion. Papers with many citations commonly validate observed effects from several angles and unravel the underlying mechanisms.
What is the cause for the observed effect? Just because a substance causes an effect and a change in a protein along a specific pathway is observed does not necessarily mean that this protein is causally involved in initiating the effect. In contrast, the effect could induce changes in the protein as part of a compensation mechanism. Although this might seem obvious, the question of whether compounds like resveratrol are anti- or pro-oxidative remains unclear. Here is more on this debate, as well as the positive effects of small amounts of oxidative stress (hormesis).
Therefore, other groups can safely rely on observations and identify new targets instead of establishing the same workflows as described in the original publication to unravel mechanisms and fill gaps. Even when developing treatments, assuming the most likely mechanism can lead to dead ends if it turns out to be incorrect.
Technical Validity vs Relevant Validity
Great papers avoid the wasteful copying of workflows because they do not leave many gaps to their validity. How to copy it?
If a Western blot shows increased protein concentration after treatment, one should not follow up with a Bradford assay or UV absorption measurement. These are easy to perform and provide technical validation, but they add little meaningful information.
It is better to unravel the mechanism behind the changes as validation. For example, one can use CRISPR or other transfection techniques to artificially increase the concentration of the protein in question.
Then, it is to be tested whether this increase replicates the effect one originally observed. This approach allows to determine whether the change in protein concentration is merely correlated with the observed outcome or actually causally related.
Adding Information
The next step is to further understand the mechanism. Therefore, one must be deliberate in distinguishing between upstream, same-level, and downstream factors.
After observing changes in JNK, it is less interesting to investigate P38 than ERK becaues the latter is part of another pathway. Of course, the pathway is a bit more complex, as MKK3/6 can affect P38 and I also left MKK7 out. You can read more here.
Let us assume once more taht you observed a higher concentration of your target protein. Now you could investigate concentration changes of proteins on the same level i.e., affected by the same transcription factor.
However, you would not provide a better understanding of the underlying processes. In contrast, you could show which changes (e.g., phosphorylation or confirmation change) in the transcription factor cause the changes in transcription.
Next step: think about upstream or downstream targets instead of jumping ahead to test whether a natural compound can reverse these concentration changes.
If you observe a higher protein concentration, the easiest way to extend your findings is to perform a qPCR on the associated gene—this helps determine whether the change is due to increased mRNA levels or reduced protein degradation.
Not Building Islands
However, too often, even strong studies focus on a question without properly building on previous research. Unfortunately less a problem for publishing, yet for scientific progress and sustainability it is.
Note that you are neither able to draw a conclusion (the middle point is not covered), nor can you compare findings in Paper 1 and 4 (no overlap)
For example, a paper might show increasing phase separation (or protein confirmation changes if you like better) at 20nM, 35 nm and 50nM but not no changes at the 2 nM control concentration. The next question may be how quick this effect is to appear. You might want to study 1ms, 2 s, 5 s and 1min time points. However, choosing a concentration of 15nM to test these time points will certainly result in uncertainties with references to the original paper.
Moreover, if possible, use positive controls – while they help to validate your findings, they also indicate effect strength.
Finally, compare to previously described effects.
If a previous study showed that resveratrol from grape extracts has antioxidative properties and you demonstrate the same for quercetin, include resveratrol in your assay for comparison. Is quercetin more potent? If you apply both, is the effect synergistic or does it saturate? If you don’t do this test, someone else has to set up these experiments (wasting resources) and might steel your citations.
Applying The Knowledge
Avoid discussing hypotheses when you can provide proof—ask yourself, how can you unravel processes conclusively?
Just because you publish in nature, you are not guaranteed many publications. Nevertheless, the curves of PLOS ONE and Nature Communications look quite different. To create these graphs, a protocol to normalize the area under the curve to 1 was applied, resulting in the "Fraction of Total Papers." Of course, you can see an extremely high data point at 100 once again—these publications strongly influence the IF. As a result, the calculated number of publications with over 100 citations underestimates the actual number.
The goal is to pin down the mechanism of change. Don’t just claim that grape extract increases protein levels—show that it alters epigenetic mechanisms, activating a novel downstream pathway, which, in turn, suggests that optimizing existing drugs won’t improve outcomes.
Nature papers typically connect protein-level changes with genetic-level effects and behavioral changes in mouse models. But what if you don’t have the resources?
Perform a qPCR and investigate changes in cell cultures, migration assays, or organoids. You might not publish in Nature, but your work will likely land in a higher-impact journal than you originally expected and save a lot of resources in other labs.
Upcoming Lesson:
How To Advocate Without Endangering One's Position
How We Feel Today
References
Romanovsky, M., 2019. Distribution of scientific journals impact factor. arXiv. doi:10.48550/arXiv.1904.05320.
Larivière, V., et al., 2016. A simple proposal for the publication of journal citation distributions. bioRxiv. doi:10.1101/062109.
Cargnello, M., et al., 2011. Activation and function of the MAPKs and their substrates, the MAPK-activated protein kinases. Microbiol. Mol. Biol. Rev., 75(1), 50-83. doi:10.1128/MMBR.00031-10.
Shaito, A., et al., 2020. Potential adverse effects of resveratrol: A literature review. Int. J. Mol. Sci., 21(6), 2084. doi:10.3390/ijms21062084.
Plauth, A., et al., 2016. Hormetic shifting of redox environment by pro-oxidative resveratrol protects cells against stress. Free Radic. Biol. Med., 99, 608-622. doi:10.1016/j.freeradbiomed.2016.08.006.
de la Lastra, C.A., et al., 2007. Resveratrol as an antioxidant and pro-oxidant agent: mechanisms and clinical implications. Biochem. Soc. Trans., 35(Pt 5), 1156-1160. doi:10.1042/BST0351156.
If you have a wish or a question, feel free to reply to this Email. Otherwise, wish you a beatiful week! See you again the 20th : )
Personal Note from Patrick, the Editor Hi Reader, there is a lot of innovation happening in green labs. Last week, we held our summit, with some of the biggest companies worldwide participating. What became very clear to me is that change is becoming easier than ever before. Here are my 3 key takeaways from the event, as well as the recording: Today's Lesson: The State of Innovation Where is the green labs field heading? Number of the Day Over the past five years, I have collected resources...
Personal Note from Patrick, the Editor Hi Reader, there is one very well-known company that is lying to you. I am talking about manipulating data and greenwashing. More about that later today. Here is your link to join. But to stay positive, today you will also hear from a scientist turned entrepreneur. Dr. Katie Day turned her passion into a company that helps prevent bovine fetuses from being slaughtered for FBS. The advantage? The alternative to FBS is chemically-defined and more reliable....
Personal Note from Patrick, the Editor Hi Reader, today and tomorrow are big days! First, here is the line-up, and here is the link for you to join. Why are we so excited? Because we believe we can bring you something truly unique. Let me explain and share a few news: Greener LabsDay #1 4 PM Berlin | 11 AM New YorkCheck Out Your Time Transforming Science in No Time For the first talk, I've been going through all the literature I've read over the past 5 years. Unfortunately, it is still almost...
