Skip to main content

Thank you for visiting You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.

  • NEWS

COVID in Venezuela, health trials suppressed — the week in infographics

Venezuela COVID stats

Official figures for deaths caused by COVID-19 in Venezuela are significantly lower than those reported in neighbouring countries such as Brazil and Colombia. Venezuela’s parlous economic state could be hampering transmission of the virus — but government data do not line up with reports collected at hospitals, indicating that the government might be under-reporting cases. For example, Médicos Unidos Venezuela, a collective of doctors monitoring the situation, sometimes reports higher weekly death rates for Venezuela’s health-care workers than the government reports for the entire country.

QUESTIONABLE COVID DATA. Venezuela has reported fewer cumulative COVID-19 deaths leading to accusations of inaccurate counting.

Source: Our World in Data

Funder pressure

A survey of public-health researchers has found numerous instances of trial results being suppressed on topics such as nutrition, sexual health, physical activity and substance use, with 18% of respondents reporting that they had, on at least one occasion, felt pressure from funders to delay reporting, or to alter or not publish findings. In the survey, published in PLOS ONE, respondents noted that they were more likely to report pressure from government department funders seeking to influence research outcomes than from industry or charity funders, or public research funding agencies.

HOW TRIAL FINDINGS WERE SUPPRESSED. Graphic showing ways funders attempted to interfere in the publication of trials.

Source: McCrabb, S. et al. PLoS ONE 16, e0255704 (2021).

Confidence and AlphaFold2

The machine-learning system AlphaFold2 has been used to predict the structures of all human proteins that self-assemble into specific 3D structures. Importantly, its predictions are accurate enough to generate biological insights and hypotheses that can be tested experimentally. The system even offers calibrated self-assessment, giving a reliable estimate of correctness at the level of individual amino-acid residues. Using a confidence metric called the predicted local distance difference test (pLDDT), it estimated how well the predicted position of each amino-acid residue agreed with experimentally determined positions for residues that were previously resolved in experiments (3,440,359 residues); residues that could not be resolved in experiments (589,079 residues); and all of the residues in human proteins (10,537,122 residues).

The confidence of protein-structure predictions by AlphaFold2



Nature Careers


Nature Briefing

Sign up for the Nature Briefing newsletter — what matters in science, free to your inbox daily.

Get the most important science stories of the day, free in your inbox. Sign up for Nature Briefing


Quick links