Research articles

Genome-wide association analyses identify risk loci for spontaneous coronary artery dissection and implicate arterial integrity and tissue-mediated coagulation in the disease etiology. Several risk variants show opposite effects on coronary artery disease risk.

Analysis of rare coding variants in the UK Biobank identifies eight genes associated with adult cognitive function, including KDM5B. Rare and common variant signals overlap and contribute additively to the phenotype.

Analyses of whole-genome and RNA sequencing data from 2,733 African American, Puerto Rican and Mexican American individuals reveal ancestry-specific patterns in the genetic architecture of whole-blood gene expression.

Single-cell multiomic and functional characterization of human pancreatic islets identifies two beta cell subtypes correlated with type 2 diabetes progression that exhibit distinct gene regulatory programs and electrophysiological phenotypes.

The resistance gene Lr9, which was introduced into bread wheat from the wild grass species Aegilops umbellulata, encodes an unusual tandem kinase fusion protein that confers wheat leaf rust resistance.

The resistance gene Sr43, which was crossed into bread wheat from the wild grass Thinopyrum elongatum, encodes an unusual protein kinase fusion protein that confers wheat stem rust resistance.

Causal robust mapping method in meta-analysis (CARMA) studies incorporates flexible prior distributions, joint modeling of summary statistics and functional annotations and outlier detection for improved causal variant fine-mapping in genome-wide association meta-analyses.

MEGAnE is a new tool to detect and genotype mobile element variants (MEVs) from short-read whole-genome sequencing datasets. Genetic analyses implicate MEVs as population-specific drivers of gene expression variation and disease risk.

Genomic and transcriptomic analysis of 470 mostly high-risk neuroblastomas collected from 283 patients delineates subtype-specific evolutionary patterns and progression-related convergent evolution and describes the clonal dynamics of metastases.

A pan-cancer analysis of primary and metastatic tumors highlights the diversity of genetic immune escape mechanisms established during tumor evolution. The authors also present LILAC, a tool to characterize the HLA-I locus from whole-genome sequencing data.

Genome-wide association analyses across individuals of East Asian and European ancestries identify new risk loci for inflammatory bowel diseases. A polygenic risk score derived from the combined datasets shows improved prediction accuracy.

Region Capture Micro-C (RCMC) combines MNase-based 3C with a tiling region-capture method. Profiling mouse embryonic stem cells with RCMC identifies nested microcompartments, which connect enhancers and promoters.

scEC&T-seq profiles extrachromosomal circular DNA and full-length mRNA from single human cancer cells, and may be used to interrogate heterogeneity in both cell lines and primary tumor samples.

AbSplice predicts aberrant splicing for 50 human tissues by integrating sequence-based deep learning models, DNA variation and RNA-seq obtained from accessible tissues.

ARG-Needle is a method to infer genome-wide genealogies from large-scale genotyping data that can be used in association analyses. Applied to UK Biobank data, genealogy-based testing finds more trait associations than using imputed genotypes.

Concatenating Original Duplex for Error Correction (CODEC) is a method that concatenates both strands of each DNA duplex to enable highly sensitive mutation detection in a range of analytes with fewer reads and lower error rates than current methods.

Peripheral blood mononuclear cells from 73 Japanese patients with coronavirus disease 2019 (COVID-19) and 75 healthy controls were analyzed using single-cell transcriptomics. Combining these data with genotyping data highlights the interplay between host genetics and the immune response in modulating disease severity.

Genome-wide association analyses identify 11 loci associated with native myocardial T1 time, a marker of interstitial fibrosis, providing insights into the pathways involved in myocardial fibrosis and myofibroblast cell state acquisition.

A deep convolutional neural network calculates liability scores for chronic obstructive pulmonary disease (COPD) from raw spirogram traces and noisy medical-record-based labels in the UK Biobank. Genome-wide analyses using these scores replicate known loci for lung function and identify 67 new disease loci.

‘MAximum Parsimonious Likelihood Estimation’ (MAPLE) is a maximum likelihood-based approach for inference of phylogenetic trees from very large datasets of similar sequences incorporating a sparse alignment representation and parsimony-based approximations, offering higher accuracy and reduced computational requirements.