Science Paper Trends You Need to Watch

With so much innovative and fascinating research, sometimes deciding which science headlines to keep an eye on can be overwhelming. Here are two trends in science articles that offer some insight into future successful papers.

When to Publish

It stands to reason that authors of scientific papers, those published in peer reviewed journals, are more likely to publish a pivotal piece as their career progresses. After all, it is reasonable to think that their research questions would be refined as time passes and experience would lead to better experiment concepts, design, and execution. This reasoning, it turns out, is not supported empirically. Philip Ball, in Scientists Can Publish Their Best Work at Any Age, published in the journal Nature (Nov 3, 2016), reports this hypothesis unsubstantiated. Albert-László Barabási and colleagues tested the hypothesis that subsequent papers had a greater likelihood of becoming breakthroughs. “‘We scientists are random,’ Barabási says. ‘Every time we publish a paper, we have the same chance of publishing our biggest hit as we do with any other paper.'” Not only was publication success unrelated to career phase, it was related to an innate (maybe) and unchanging “Q factor”.  Success “depends on only two factors, they argue: an element of luck, and a certain quality, or Q factor, that measures an individual scientist’s ability to boost the impact of any project.” Q factor seems to reflect scientist education and communication skills.

Take away: when looking for upcoming influential papers, focus on upcoming papers from well educated, persuasive authors rather than those with greater seniority.

Informed Estimates

The human microbiome may just be our overlooked organ, variable in composition, touching virtually every physiological function. For nearly 4 decades, science widely held that for each of the some 3 trillion human cells in the standard human there are 10 times as many bacterial cells (more accurately microbial cells, to include fungi, protist, and archaea in addition to bacteria). Sender and colleagues used a method of analysis popularly called Fermi problems to test this. The importance of the paper published in PLoS Biology (Aug 19, 2016), although the estimate of the human:microbial cells is important, is the method they used to reach their estimate. A Fermi problem holds that an unknown value can be estimated by drawing upon knowledge of underlying properties of the unknown value. For example, the estimated weight of an apple might be ascertained through a Fermi problem: while the weight might not be known, the weight of other common objects are known. A box (four sticks) of butter weighs one pound, and is probably on the order of magnitude of an apple, maybe twice the size, although butter might be slightly less dense than an apple, some ‘wiggle room’ will be added into the estimate. Since one apple is thought to be half the size of a box of butter, the starting estimate is one half pound; to estimate the range of possible weights, one estimates the minimum plausible uncertainty of the estimate. I’ll use 50% as the uncertainty because the butter’s density is likely to be greater than half that of the apple, and the apple’s actual size is not likely twice that of the perceived size. This yields an estimated range of 1/4 pound to 1 pound. A typical medium apple weighs just over six ounces, or squarely within the uncertainty range. Sender et al used a similar approach to estimating the human:microbial cells in the body.

Take away: estimates provide an essential reference point in drawing predictions from hypothesis and in designing experiments. Look for innovative, and increasingly accurate estimation techniques in scientific papers as a sign of those with high Q factor and likely to have more accurate, results.