Genetics — Weekly Evidence Brief

The Causal Pivot (CP) model offers a structural approach to analyze genetic heterogeneity, significantly improving the discovery of causal genetic variants in complex diseases like hypercholesterolemia, breast cancer, and Parkinson's disease 1. This method leverages polygenic risk scores and evaluates rare variants, holding potential for future diagnostics. Advanced computational tools are also refining variant detection; TRsv, a new tool, can simultaneously detect tandem repeat copy number variations (TR-CNVs), structural variations (SVs), and short indels from long-read sequencing data 2. Similarly, a novel deep learning approach significantly enhances accurate calling of indels in family trios from long-read data, which is crucial for understanding disease pathogenesis 3. These technological advancements are vital for precision medicine, though caution is needed, as assumptions about variant effects, like premature termination codons, may not always hold true 4.

Extrachromosomal DNA (ecDNA) is recognized as a critical driver of tumorigenesis, oncogene amplification, and tumor heterogeneity, directly impacting drug resistance 5. These circular DNA molecules form 'ecDNA hubs' that amplify oncogene expression, making ecDNA a promising therapeutic target and diagnostic biomarker. Beyond ecDNA, alterations in 3D genome structures are observed during cancer progression, distinguishing morphological cancer states and highlighting their potential as diagnostic and prognostic biomarkers 6. Furthermore, pro-tumor enhancers and enhancer connectomes, identified through CRISPR screening, reveal how risk-associated single-nucleotide polymorphisms promote glioma progression by regulating genes like SOX18 7. Studies also indicate that metastatic lesions in prostate cancer evolve independently, with most mutations arising after spread and confined to individual sites, impacting diagnostic and targeted therapy strategies 8.

Genetic studies are shedding light on the molecular underpinnings of complex diseases. The largest genetic study of reading-related phenotypes to date identified 80 independent genome-wide significant loci for dyslexia and quantitative reading skill, revealing enrichments in neuronal biological process pathways 9. This multivariate genome-wide association study (GWAS) significantly boosts power for gene discovery. Beyond SNPs, RNA splicing changes are emerging as crucial in disease; a large transcriptomic study identified RNA splicing and transcript isoform-level changes in endometrial tissue linked to endometriosis, with specific splicing quantitative trait loci (sQTLs) associated with disease risk for genes like GREB1 and WASHC3 10. The architecture of regulatory networks is also critical for complex traits, with heritability clustered in tissue-specific "hubs" that are preferential targets for selection, offering a framework for understanding trait evolution and adaptation 11.

The mammalian mitochondrial genome copy number (mtCN) varies extensively across human and murine tissues, with high mtCN correlating with resistance to GPX4 loss, revealing broad biological implications 12. Transposable elements (TEs) exhibit specific expression patterns and subcellular dynamics during human pluripotent stem cell differentiation, suggesting a role in human embryonic development through dynamic chromatin interactions 13. While TEs can be silenced by piRNA clusters, some demonstrate a biased insertion into these regions, altering invasion dynamics 14. These fundamental genomic mechanisms highlight the intricate layers of regulation contributing to both normal development and disease pathogenesis.