Genomic language models: opportunities and challenges

Trends in Genetics, Год журнала: 2025, Номер unknown

Опубликована: Янв. 1, 2025

Язык: Английский

---

Transformers and genome language models

Nature Machine Intelligence, Год журнала: 2025, Номер unknown

Опубликована: Март 13, 2025

---

Identification of Novel Mosaic Variants in Focal Epilepsy-Associated Patients’ Brain Lesions

Genes, Год журнала: 2025, Номер 16(4), С. 421 - 421

Опубликована: Март 31, 2025

Focal cortical dysplasia type III (FCDIII) is a rare and complex condition associated with drug-resistant epilepsy often characterized by lamination abnormalities, along variety of neoplasms vascular abnormalities. Objectives: This study aimed to elucidate the genetic architecture underlying FCDIII through use whole-exome sequencing (WES) brain peripheral blood samples from 19 patients who had been diagnosed FCDIII. Methods: Variants were identified series machine-learning-based detection functional prediction methods not previously Mosaic fraction scores these variants validated variants’ pathogenicity, in silico gene ontology enrichment analyses demonstrated that severe destabilizing effects on protein structure. Results: We reported ten novel pathogenic somatic missense loss function across eight genes, including CNTNAP2, ACY1, SERAC1, BRAF. Genetic alterations linked clinical manifestations, such as encephalopathies intellectual disabilities, thereby emphasizing their role molecular drivers Conclusions: next-generation sequencing-based mosaic variant-calling pipelines are useful for diagnosis FCDIII, opening up avenues targeted therapies, yet further research required validate findings examine therapeutic implications.

Язык: Английский

---

Genome modeling and design across all domains of life with Evo 2

bioRxiv (Cold Spring Harbor Laboratory), Год журнала: 2025, Номер unknown

Опубликована: Фев. 21, 2025

Abstract All of life encodes information with DNA. While tools for sequencing, synthesis, and editing genomic code have transformed biological research, intelligently composing new systems would also require a deep understanding the immense complexity encoded by genomes. We introduce Evo 2, foundation model trained on 9.3 trillion DNA base pairs from highly curated atlas spanning all domains life. train 2 7B 40B parameters to an unprecedented 1 million token context window single-nucleotide resolution. learns sequence alone accurately predict functional impacts genetic variation—from noncoding pathogenic mutations clinically significant BRCA1 variants—without task-specific finetuning. Applying mechanistic interpretability analyses, we reveal that autonomously breadth features, including exon–intron boundaries, transcription factor binding sites, protein structural elements, prophage regions. Beyond its predictive capabilities, generates mitochondrial, prokaryotic, eukaryotic sequences at genome scale greater naturalness coherence than previous methods. Guiding via inference-time search enables controllable generation epigenomic structure, which demonstrate first scaling results in biology. make fully open, parameters, training code, inference OpenGenome2 dataset, accelerate exploration design complexity.

Язык: Английский

---

Genomic language models could transform medicine but not yet

npj Digital Medicine, Год журнала: 2025, Номер 8(1)

Опубликована: Апрель 18, 2025

Язык: Английский

---