Shamil R. Sunyaev

The paper introduces KASM to integrate functional annotations and estimate heritability of rare protein coding variants across 786,871 exomes, finding higher heritability than previous estimates.

The paper identifies domain-specific phenotypic effects of MACF1 variants outside the GAR domain, revealing broader neurodevelopmental and craniofacial anomalies.

NERINE tests gene network hypotheses with rare variant associations, identifying a subnetwork linked to Parkinson's disease.

This paper demonstrates that ensilication preserves DNA integrity at ambient temperature, enabling diagnostic-quality long-read sequencing without cold-chain infrastructure.

The paper estimates mutation rates and variant effects using 700k exomes to identify genes with mutational hotspots and improve variant prioritization for genetic diagnostics.

gnomAD v4 (730,947 exomes) brings LOFTEE-2 loss-of-function annotation (0.90 precision, strict mode), missense-aware constraint, and LLM-mined literature evidence combined with constraint to rank disease genes.

The paper investigates missing regulation by comparing trait-associated variants in cattle and pigs to human genetics, finding increased regulatory variant ascertainment in livestock.

The paper analyzes the distribution of simple repeat lengths across mammalian genomes and shows that long repeats are abundant due to mutational dynamics, not natural selection.

The paper introduces scJLIM, a cluster-free method for single-cell colocalization testing, improving power to link GWAS loci with gene expression changes in specific cell types.

The paper identifies genes with hypermutable de novo mutations as potential drivers of clonal expansions in spermatogonia, highlighting their impact on mutation rates and disease associations.

The paper uses DNA lesion segregation to estimate tumor initiating events and shows early cancer drivers have a fitness advantage exceeding 30%.

The paper develops a mitochondrial constraint model applied to gnomAD data, revealing strong depletion of expected variation and identifying constrained regions in mtDNA.

The paper analyzes GWAS data for 27 traits to test evolutionary models, finding most traits are consistent with stabilizing selection rather than neutral evolution or directional selection.

The paper identifies a shared genetic risk module involving PSEN2 and actin regulation across alpha-synucleinopathies, linking them to Alzheimer's disease mechanisms.

RExPRT predicts pathogenicity of tandem repeat expansions with high precision, identifying ~30,000 likely pathogenic TRs in the reference genome.

The paper introduces LDSPEC to estimate correlations between causal disease effects of proximal SNPs, finding significant correlations that vary with distance, allele frequency, LD, and functional annotations, implicating stabilizing selection.

The paper introduces a basepair-resolution mutation rate model, Roulette, which identifies a 10-fold increase in mutation rate for Polymerase III-transcribed genes and detects elevated rates in testis-specific transcription factor binding sites.

The paper uses primate genetic data to predict human variant pathogenicity with a deep learning classifier trained on 4.3 million common missense variants.

This paper uses pleiotropy to discover new genetic loci for OSA and provide biological hypotheses by combining colocalization tests with pleiotropy analysis.

The paper investigates the link between genetic associations and regulatory function, finding limited evidence that baseline gene expression explains GWAS associations.

FAVOR provides an online resource and tool for functional annotation of variants from whole genome sequencing studies to aid in disease gene prioritization.

AnFiSA is an open-source platform for analyzing sequencing data in rare genetic disease, addressing reproducibility, adaptability, and explainability in genomic analysis.

This study identifies 114 genome-wide significant rare CNV-disease associations across 30 neurological disorders, revealing 38 novel triplosensitive genes and developing a model to predict dosage sensitivity.

The paper identifies rare variants in the complement system associated with severe sepsis with coagulopathy using pathway-based analysis of whole exome sequencing data.

The paper identifies prenatal mutational processes in schizophrenia brains linked to transcription factor binding sites and proposes a new term 'skiagenesis' for these mutations.

The paper analyzes mutagenesis at non-B DNA motifs in the human genome, revealing biological mechanisms and correcting previous findings by accounting for confounding factors.

StrVCTVRE predicts the pathogenicity of human genome structural variants using a random forest classifier, achieving 90% sensitivity while filtering half of SVs.

This study develops a method to detect recessive purifying selection in genes using population genetics simulations and applies it to empirical gene sets, finding significant enrichments for recessive selection in specific gene categories.

The paper improves genetic mapping resolution in autoimmune diseases by identifying shared alleles across multiple conditions and reveals widespread shared mechanisms.