Thank you for visiting nature.com. 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.
Clarke, Evangelista et al. use a knowledge graph (ReproTox-KG) to characterize associations between small molecule compounds and their potential to induce specific birth abnormalities. They identify over 500 birth defect/gene/drug connections that can explain molecular mechanisms for drug-induced birth defects.
Ganesan et al. examine electronic health record data from over 100,000 patients with Dupuytren’s disease, a condition characterized by contracture of the hand, to look at its association with diabetes. They report a higher prevalence of Dupuytren’s disease in type 2 than in type 1 diabetes and in patients with HbA1c levels within the diabetic range.
Jimenez-Silva, Rivero et al. perform a genomic epidemiology analysis of SARS-CoV-2 variants in Colombia. They show that in the first two years of the pandemic a total of 188 SARS-CoV-2 lineages have been found circulating in the country, including the Mu lineage that caused a wave of infections not seen in other countries.
Ensenyat–Mendez et al. construct a gene expression-based machine learning classifier to predict the response of triple-negative breast cancer to immune checkpoint inhibition combined with chemotherapy. Predictive performance of the 37-gene classifier is better than that of PD-1 or PD-L1.
Liu et al. develop an artificial intelligence approach to automatically generate radiological reports for acute stroke MRIs. The system quantifies ischemic infarcts and reports their location, is publicly available, and runs with minimal computational requirements.
Holm, Ivarsdottir, Olafsdottir et al. compare symptoms and physical measures between Icelanders post SARS-CoV-2 infection with uninfected controls. From reported symptoms, they estimate the prevalence of long COVID as 7% at a median of 8 months after infection, while objective differences between cases and controls in the physical measures were few.
Qasmieh et al. estimate SARS-CoV-2 prevalence and the frequency of related public health outcomes in New York City in April/May 2022, via a cross-sectional survey of 1030 adults. Their findings suggest SARS-CoV-2 prevalence during the BA.2/BA.2.12.1 surge in New York City may have been underestimated by routine surveillance.
Arumugam, Ma et al. automate the interpretation of lateral flow assays by applying their software architecture to photos taken with a smartphone camera. The approach is rapidly adapted to other test kits using only a fraction of the images required for existing methods, with all test results successfully classified.
Lin et al. report an association between increased circulating endothelin-1 levels and higher mortality risk in patients with stable coronary artery disease. High intensity statin therapy is associated with a reduced risk of all-cause mortality or cardiovascular death in patients with high endothelin-1 levels.
Riemer et al. explore the relationship between early microstructural changes in subcortical volumes and development of depression in patients with relapsing-remitting Multiple Sclerosis (RRMS). Higher levels of free-water in the subcortical structures at an early stage of RRMS are associated with depression symptoms at a later stage of the disease.
George, Ellis et al. develop an explainable artificial intelligence model to predict 30-day emergency department admission in patients with cancer undergoing treatment. The authors describe an approach to monitor the model after deployment, which provides ongoing readouts of data consistency and model behavior.
Pisharady et al. demonstrate changes in diffusion metrics and cortical thickness in brain and cervical cord MRI in people with Amyotrophic lateral sclerosis (ALS) over time. Fiber density and fiber cross-section show the largest effect and may therefore be promising imaging biomarkers.
Kikuchi, Michikawa et al. investigate whether sleep quality and temperament in 1-month-old infants is associated with Autism Spectrum Disorder (ASD) in 3-year-old children in Japan. Infants who had longer daytime sleeps and more intense crying had a higher risk of ASD at age 3.
Pallett, Heskin et al. monitor longitudinal response in older adults following vaccination with the BNT162b2 COVID-19 (Pfizer-BioNTech) vaccination. Hybrid immunity is associated with higher antibody titres, neutralisation and inhibition capacity.
Paranjpe, Jayaraman et al. profile the plasma proteome to identify markers of acute and long-term COVID-19-associated kidney dysfunction. They find that while both acute and long-term kidney dysfunction are associated with tubular dysfunction markers, the acute phase is also associated with markers of myocardial damage and hemodynamic disturbance.
Koher et al. use Bayesian analysis to evaluate the usefulness of survey responses compared to mobility data as a tool to monitor the effects of lockdown as a disease mitigation strategy. Self-reported contacts during lockdown better predict future hospitalizations than mobility data.
Oyake et al. survey patients with neurological and orthopedic disorders undergoing inpatient rehabilitation and the medical practitioners treating them about motivational factors. Whilst there are some differences, the most frequent motivational factors are shared across both patients and medical practitioners.
Bah, Kujabi et al. characterise multi-drug resistant-Gram negative bacilli (MDR-GNB) carriage in small vulnerable newborns and their mothers at a low-resource African hospital. Many neonates acquire MDR and ESBL-GNB between birth and 7 days of age, but there is only limited evidence these are transmitted from the mothers.
Borah et al. develop a rapid digital pathology approach for the analysis of fresh tissue specimens, avoiding the need for frozen section histopathology. They apply the method to fresh human brain specimens and demonstrate their ability to distinguish normal and tumor regions.
Hudock et al. forecast the future impact of metastatic cancer in the USA. They predict by 2040 there will be decreased incidence of lung cancer, increased incidence of highly screened cancers, such as breast cancer, and greater odds of long-term survival.