Search

Polygenic scores for body mass index and obesity constructed using data from 5 million people with different ancestries were associated with distinct weight gain trajectories from early childhood to adulthood.

Shah and colleagues used machine learning to integrate features from computed tomography scan images, programmed death ligand-1 (PD-L1) immunohistochemistry and genomics into a multimodal predictor of response to anti-PD(L)1 in patients with NSCLC that outperformed single features such as tumor mutational burden.

In a large, partially prospective cohort of patients with molecularly profiled and clinically annotated meningioma, the extent of surgical resection and radiotherapy (RT) response correlate with molecular classification, which can be used in a molecular model to predict clinical outcomes in response to RT.

A method, RARE-seq, for sensitive detection of cell-free RNAs in blood is demonstrated to have diverse clinical applications including diagnosing and characterizing human cancers, and tracking response to RNA therapeutics.

Integrated analyses reveal that multiomics captured the heterogeneity of metabolic states accompanying obesity and changes in metabolic health in response to lifestyle intervention that are not apparent in body mass index measurements.

Chest computed tomography (CT) is one of the most common diagnostic tests. Here, the authors combine two AI models to measure from CT coronary artery calcium, left ventricular mass index, and left and right atrial and ventricular volumes, and show their association with cardiovascular mortality.

Large-scale sequencing efforts have uncovered a large number of secondary metabolic pathways, but the chemicals they synthesise remain unknown. Here the authors present PRISM 4, which predicts the chemical structures encoded by microbial genome sequences, including all classes of bacterial antibiotics in clinical use.

The authors show that rare genetic variants contribute to large gene expression changes across diverse human tissues and provide an integrative method for interpretation of rare variants in individual genomes.

Multitrait genome-wide analysis of glaucoma and related phenotypes identifies new risk loci and enables development of a polygenic risk score to predict disease susceptibility and key clinical outcomes.

A machine learning model predicts on-target and off-target activity of Cas13d in human cells.

Kruper et al. investigate the impact of glaucoma on brain tissue by comparing MRI data from 856 people with glaucoma to matched controls without glaucoma. They find that neural networks using tissue properties from visual brain connections more accurately detect glaucoma than those trained on non-visual brain connections.

Multi-omics datasets are integrated to generate a unified and clinically informed molecular classification of meningiomas.

REM-sleep behavior disorder often precedes Parkinson’s disease or dementia. Here, the authors perform a genome-wide association study for REM-sleep behavior disorder, and discover how it potentially affects gene expression in the brain.

Biller et al. use long-read sequencing of fractionated ocean nanoparticles, showing that extracellular vesicles and virus-like particles contain distinct pools of genetic information and thus may differentially contribute to horizontal gene transfer dynamics in the ocean.

The app-based COVID Symptom Study was launched in Sweden in April 2020 to contribute to real-time COVID-19 surveillance using daily symptom reports from study participants. Here, the authors show how syndromic surveillance can be used to estimate regional COVID-19 prevalence and to predict later COVID-19 hospital admissions.

DrugnomeAI predicts the druggability likelihood for every protein-coding gene in the human exome by small molecules, monoclonal antibodies, and proteolysis-targeting chimeras (PROTACs).

A trans-ancestry genome-wide association study of serum urate levels identifies 183 loci influencing this trait. Enrichment analyses, fine-mapping and colocalization with gene expression in 47 tissues implicate the kidney and liver as key target organs and prioritize potential causal genes.

A wealth of gene expression data is publicly available, yet is little use without additional human curation. Ma’ayan and colleagues report a crowdsourcing project involving over 70 participants to annotate and analyse thousands of human disease-related gene expression datasets.

Sinking of organic particles to the deep seafloor is fundamental to ocean carbon cycling. Here, the authors investigate prokaryotic communities in sinking and suspended particles, identifying depth-specific signatures of particle export and carbon cycling processes.

Perry et al. demonstrate that regulatory T (T_reg) cell function is restrained by the cell-autonomous action of the checkpoint inhibitor molecule PD-1. This PD-1-dependent mechanism tunes T_reg cell function during homeostasis and infection.

Tom Karlsen and colleagues report an association study for primary sclerosing cholangitis (PSC), a severe liver disease, using the Immunochip array. They identify nine loci newly associated with PSC and examine pleiotropy with other autoimmune disorders.

Detection of circulating tumor DNA using MRD-EDGE, a machine-learning-guided single-nucleotide variant and copy-number variant detection platform for signal enrichment, enables monitoring of minimal residual disease and immunotherapy response in settings of low tumor burden.

Global temperature fluctuations during the last 2,000 years caused consistent changes in ocean evaporation and atmospheric moisture condensation processes, reflected in coherent water isotope signals in a large compilation of proxy records.

Thomas et al. test 6 ELISAs detecting IgA and IgG antibodies to whole SARS-CoV-2 spike protein, to its receptor binding domain region and to nucleocapsid protein in saliva. Across 20 household outbreaks, antibody responses are heterogeneous, but a reliable indicator of recent infection.

Mendelsohn et al. evaluated in a prospective multicentre validation study whether host-blood transcriptomic tuberculosis (TB) signatures could be used to find active cases of TB. Whilst most of the signatures met WHO Target Product Profile criteria for a triage test to diagnose symptomatic TB; most signatures did not meet the criteria for asymptomatic TB.