Meta-analyses are often conducted to increase the generalizability of results from individual studies and can shed light of the reproducibility of a given effect. However, in an article published in January this year, Maassen and collaborators [1] note concerns about the reproducibility of these meta-analytic methods. They find that almost half of 500 primary study effect sizes drawn from 33 published psychological meta-analyses could not be reproduced. One of the tools often used to assess reproducibility may not be as reproducible as hoped.

Craig-Atkins and colleagues [2] use data from human remains, pottery shards, and animal bones to track dietary changes after the Norman conquest of England. They find that short-term food insecurity was followed by changes in preferences and intensification of the economy.

Gabriel Schwartz and Jaquelyn Jahn [3] analyzed data from Fatal Encounters—the most comprehensive database tracking police-involved deaths in the United States—and uncovered vast differences across metropolitan statistical areas (MSAs) in the overall rates of persons killed by police officers, regional differences (West and South v. Midwest and Northeast), as well as racial inequities in those rates, between 2013 and 2017, for 5,494 police-related fatalities. A formal comment by Nix [4] on this article emphasized the difficulties inherent in such research, such as limitations in the database, and the challenges in interpreting these results.

In this Registered Report Protocol, one of the first ones published at PLOS ONE, Jaffé and Douneva [5] describe three proposed studies to determine how sharing secrets affects the way individuals think about each other. We already know that closeness can make secret-sharing more likely, but these authors plan to investigate whether secret-sharing might in turn increase perceptions of closeness. These studies are expected to be conducted on online platforms and in person with students over this academic year.

Deep learning is emerging as a potential tool for a range of applications in cancer. Predicting patient prognosis can help to inform monitoring and treatment strategies. In their study, Wulczyn and colleagues [6] demonstrate the feasibility of deep learning systems for predicting survival in multiple cancer types, using histopatholgy images.

When it comes to fighting the global spread of COVID-19, the implementation of preventative measures such as social distancing are paramount. However, it can be challenging to estimate how much of a difference these measures make and how this varies depending on when these measures are put in place or lifted. In their study, Ohsawa and Tsubokura [7] simulate the spread of virus infection using artificial human networks and test how contact between people in different clusters effects the timing and intensity of viral spread. The authors find that releasing a previously imposed constraint too early can lead to a “second wave” of infection that is more intense than the previous wave, and caution against lifting restrictions on social contact too soon.

Ageism’s effects on health may be compounded by individual and structural biases against the elderly. In a systematic review, Chang and colleagues’ [8] synthesis of data from 422 studies showed ageism has a global reach.

Using historical data, this paper looks at gender equity in academic surgical specialties over the past forty years. Linscheid and colleagues [9] present an analysis suggesting that more work needs to be done to ensure female surgical students and trainees are supported to ensure representation across specialties.

The epidemiological study by Gosiker and colleagues [10], which is part of Primary and Secondary Prevention of Cardiovascular Disease (CVD) Collection  launched in December this year, exemplifies the importance of carefully assessing CVD risk in vulnerable populations. Using data from a large collaboration of clinical sites in the United States, the investigators examined the prevalence of elevated 10-year CVD risk among transgender women with HIV. The approach and main findings from this study contribute to the growing need of calibrated risk prediction tools that support CVD preventive strategies in underrepresented and vulnerable populations.

The impact of socioeconomic factors (for example, education level) on health outcomes is widely studied—but do inequality rates vary across different healthcare systems types? Rydland and colleagues [11] contribute to the debate by analysing how educational inequalities found in mortality amenable to healthcare (defined as deaths which are preventable through medical intervention) vary among European countries with different typologies of health systems.

Doshi and colleagues [12] interview providers to analyze the experiences of undocumented Latino(a)/Latinx immigrants seeking healthcare and social services in the United States. By framing the results according to the “Three Delays Model”, the authors provide rich insights on the lived experiences of a marginalized group, and suggest several approaches to improve access to care.

Mental health is a leading cause of disability globally. In their article, Smith and colleagues [13] showcase the use of Implementation Science (IS) tools/principles to deliver and evaluate scale up of much-needed mental health services in the Global South. This is one of a few IS hybrid studies that looked at both: a) implementation outcomes (reach/uptake of the intervention, fidelity) and b) effectiveness outcomes (clinical outcomes, functionality). The authors used elements of the RE-AIM framework/theory and applied simple quasi-experimental design methods for quantitative evaluation. 

Whale sharks, the largest living fish, don’t have much to fear from predators and are known to be docile and and approchable. However, their eyes are exposed to damage from mechanical, chemical and biological hazards. This year, Tomita and colleagues [14] from Japan’s Okinawa Churashima Research Center made the surprising discovery that their retractible eyes are uniquely protected by rows of oakleaf-like denticles. This finding highlights the importance of vision in this species.

Resistin molecules are pleiotropic cytokines associated with inflammatory diseases but their distribution features in human organs remains unexplored. A new anti-human resistin antibody developed by Lin and colleagues [15] is used to explicitly detect the expression pattern of human resistin across normal human tissues. Results show that resistin is broadly distributed, but principally localized in the cytoplasmic granules of macrophages, hematopoietic precursor cells and neuronal cells. This characterization of resistin expression provides a framework for recognition and interpretation of changes in these patterns that may be associated with disease states.

The endemic Giant Senecios of the East African sky islands are an iconic example of the gigantism that has evolved convergently in a number of tropical montane plant lineages. In their new study, Tusiime and colleagues [16] genotyped 460 plants from across the region’s mountains to assess genetic structuring, biogeographic patterns, and taxonomic concepts. They found evidence for rapid recent radiation underpinned by long-distance dispersal and parallel adaptation. They also reported low levels of overall genetic diversity, emphasising the need for enhanced conservation efforts.

PARP proteins play a critical role in the early detection of DNA double-strand breaks, but as yet no structural information on double-strand break detection in the context of chromatin exists. In their study, Gaullier and colleagues [17] use cryo-EM to solve the structure of PARP2 bridging the mimicked double-stranded break of two nucleosomes (Nuc165) reconstituted in vitro with two short linker DNA overhangs. Their model illustrates that by bridging the break, PARP2 acts as a platform for recruitment of other factors involved in DNA repair such as HPF1.

Understanding long term changes in the ocean can be invaluable not only for elucidating the historical record, but for predicting future scenarios amidst a changing climate. In this study, Simon and colleagues [18] use an unprecedentedly high-resolution record of stable isotopes from a marine core retrieved off western North Iceland to characterize long-term ocean variability. Using foraminifera, they reconstruct the record for the period before instrumentation, as well as overlapping with the period of instrumental observation, and demonstrate a high degree of agreement in predicted and observed data.

Plastic pollution of the natural environment is an ongoing global crisis and plastics which end up in waterways or the ocean can travel vast distances. In this new study, Duncan and colleagues [19] designed their own “bottle tags” which mimicked the characteristics of a half-full 500ml plastic bottle. By installing open-source technology within the bottle and releasing them at various points in the Ganges river system and the Bay of Bengal, the team was able to track the journeys that their bottles made; The longest journey was 2845km over 94 days, illustrating the dispersal processes of plastic pollution.

How do you open a historical treasure chest without a key, when the lock may contain any manner of failsafes or protective mechanisms? Zikmundová and colleagues [20] tackle this problem using X-ray computed tomography (CT) scanning. Their technique enabled a detailed exploration of the lock based on a system of levers and bolts handled by a single key, located in a case on the inside of the chest lid, including the dimensions essential for manufacturing of a new key copy. Moreover, two secret protective mechanisms were revealed, as well as all the damages of the object.