GENOME BIOLOGY基因组生物学
GENOME BIOLOGY(英文缩写 GENOME BIOL),ISSN 1474-760X,eISSN 1474-760X,中文译名:基因组生物学 是一本学术期刊。本页汇总该期刊的最新影响因子、分区信息以及最新收录于 PubMed 的文献,帮助您快速了解期刊全貌。
发文量统计区间：2025-11-01 至 2026-10-31，按本站收录文献的发表日期统计。
期刊介绍
历年影响因子趋势
| JCR 数据年份 | 影响因子 | JCR 分区 |
|---|---|---|
| 2021 | 18.010 | Q1 |
| 2022 | 12.300 | Q1 |
| 2023 | 10.100 | Q1 |
| 2024 | 9.400 | Q1 |
| 2025 | 9.200 | Q1 |
GENOME BIOLOGY 最新收录文献
-
2. dicast: a machine learning method for accurate structural variant detection from short-read sequencing data.
PMID:日期:2026-09-16Structural variants are a common cause of human diseases, but their detection from short-read sequencing remains challenging, despite being the technology underlying most clinical workflows. We present dicast, a machine-learning method that scores SV calls from short-read data using alignment and genomic-context features. dicast is trained on a new multi-technology ground truth built from nine samples, with extensive manual curation. It outperforms existing short-read callers and consensus approaches, recovering substantially more true positives at high precision. We also demonstrate dicast's applicability for diagnostics, identifying all pathogenic variants in multiple disease cohorts, and 20% more candidate pathogenic deletions than consensus approaches.
-
3. Decoding the cancer microbiome: multi-omics, AI, and translational opportunities.
PMID:日期:2026-09-16Multi-omics technologies, coupled with AI technologies, have the potential to enable the systematic investigation of complex cancer microbiome biology by uncovering informative patterns and associations across complementary datasets. Here, we review existing and emerging cancer microbiome data, discuss the development, interpretation, and validation of AI models as key considerations for their integration and analysis, and provide practical suggestions for improving the reliability and biological relevance of AI-driven discoveries. We further highlight the opportunities and challenges of translating these discoveries into clinical practice, emphasizing strategies to bridge the gap between research-oriented models and clinical implementations.
-
4. Genolator enables protein function interpretation using a multimodal large language model fusing genomic and structural interpretation with natural language interaction.
PMID:日期:2026-09-16Decoding the genetic code to unveil its genome functionality is a monumental task which would greatly advance the understanding of disease mechanisms and development of targeted treatments. Although large language models (LLMs) have transformed natural language processing across diverse domains, translating the complex language of DNA into human-readable form remains challenging due to genomic data complexity and unexplored regions of the human genome. Current language models either are capable of processing natural language or the genomic code. Models fusing both aspects are largely lacking. Here we present Genolator, a multimodal large language model that integrates embeddings from DNA sequences, amino acid sequences, and protein structures with natural language queries. Fine-tuned on over 365,000 question-answer pairs generated using abstracted Gene-Ontology (GO) terms, Genolator effectively answers queries regarding protein subcellular localization, molecular function, and biological processes. Evaluation demonstrates high accuracy in confirming or denying protein function associations, outperforming baseline models such as openly available allrounder LLMs like GPT 4.1 as well as smaller domain-specific models integrating knowledge from a protein structure transformer. Explorations of Genolator's hidden states unveil a biologically and linguistically plausible organization of its learned representations. Analysis of the attention heads of the underlying language model and an ablation study provide evidence for a benefit of the multi-modal approach. Genolator enhances accessibility to genomic information by enabling natural language interaction with protein data, facilitating biological discovery, and clinical research. It represents a step towards bridging genomic code and human language through the integration of a multimodal LLM.
-
5. Population genomics, demography, and circum-Baltic connectivity of Early Medieval southwestern Finland.
PMID:日期:2026-09-15Knowledge of Early Medieval Finland (1050-1250 CE) relies primarily on archaeological evidence, as contemporary sources are scarce. The available evidence indicates two distinct cultural-economic zones: coastal and inland. Using newly generated genomic data from 34 ancient individuals alongside modern Finnish genomes, we characterise Late Iron Age and Early Medieval ancestry in southwestern Finland, reconstruct demographic patterns, and place individuals within a circum-Baltic relatedness network. Early Medieval ancestry in inland southwestern Finland was very similar to that of present-day inhabitants. Ancient coastal and inland individuals were genetically indistinguishable, whereas modern coastal populations showed substantially more Scandinavian ancestry, and less Baltic ancestry compared to their ancient counterparts. IBD (identity-by-descent) analyses also indicate a major genetic shift in the coastal zone since the Early Medieval Period. Effective population size increased throughout the study period and was ~ 13,000 by 1250 CE. IBD links between Scandinavia and Early Medieval Finland align with known archaeological connections. Furthermore, we identify IBD links between individuals from Early Medieval Finland and victims of the Kronan warship sinking. We demonstrate nearly a millenium of population continuity in the inland zone of southwestern Finland, contrasted by a large, contemporaneous genetic shift in the coastal zone. This ancestry shift corresponds with documented medieval emigration from Sweden to Finland. The regional population rapidly expanded during this time period, likely due to new agricultural practices and favourable climatic conditions. Our circum-Baltic IBD network indicates that southwestern Finland was firmly embedded into the wider, pre-modern Baltic world.
-
6. COSIGT: population-scalable genotyping of complex loci from low-coverage sequencing data using pangenome graphs.
PMID:日期:2026-09-08Pangenome graphs capture extensive structural diversity, but resolving complex loci from shallow sequencing remains challenging, particularly when samples are of low quality such as in ancient DNA. We introduce COSIGT (COsine SImilarity-based GenoTyper), which assigns diploid genotypes by matching read-depth distributions to haplotype paths via cosine similarity. Because this metric evaluates relative coverage profiles rather than absolute read counts, COSIGT substantially outperforms existing likelihood-based tools at low coverage (1-2X). We demonstrate scalability to thousands of modern and ancient genomes, enabling robust, population-scale analyses of complex variation directly from low-coverage datasets.
-
7. Large-scale benchmarking of prokaryotic annotation tools across thousands of species.
PMID:日期:2026-09-04Genome annotation is an important step in deriving functional meaning from prokaryotic sequencing data, yet systematic evaluations guiding tool selection are lacking. We present the first large-scale investigation of four prominent open-source annotation tools (Prokka, Bakta, EggNOG-mapper, and PGAP) across 156,033 diverse genomes. This includes Escherichia coli strains for baseline performance, thousands of archaea and bacteria genomes, as well as frameshifted and metagenome-assembled genomes. Bakta excels in annotating high-quality bacterial genomes, while PGAP was better for archaeal genomes and challenging bacterial assemblies, including metagenome-assembled, fragmented, or contaminated samples. For Gene Ontology annotation, PGAP consistently provides broader term coverage, whereas EggNOG-mapper offers more terms per feature. Our findings highlight tool-specific strengths crucial for selecting optimal solutions based on genome quality, taxonomy, and origin (e.g. MAGs). This study provides an evidence-based guide for users and informs future tool development.
-
8. Protein-protein interaction network architecture of human polygenic traits reveals domain-spanning connectivity and evolutionary pressures.
PMID:日期:2026-08-27Human polygenic phenotypes arise from the combined effects of many genes that interact within molecular networks. Yet, we know little about how the structure of these networks constrains or facilitates the evolution of complex traits. Here, we systematically examine the relationship between protein-protein interaction (PPI) network architecture and evolutionary signatures across 4,756 human polygenic phenotypes. We show that genes associated with polygenic phenotypes exhibit significantly higher connectivity within the global PPI network compared to matched random gene sets. Highly connected genes are enriched for immune-related biological processes, whereas genes with fewer interactions are preferentially associated with neurogenesis-related functions. Importantly, among trait-associated genes, greater network connectivity is associated with weaker selective constraint, indicating that evolutionary pressure varies according to network embedding. Together, these findings provide a systems-level framework linking molecular interaction architecture to the evolution of human polygenic traits. To support this effort, we also develop an online portal enabling researchers to generate and explore hypotheses by identifying genes that are both highly associated and highly connected across thousands of polygenic phenotypes.
-
9. Determinants of mutation susceptibility along the genome are largely invariant across human tissues.
9. 基因组中突变易感性的决定因素在人体各组织中基本保持一致PMID:日期:2026-08-22The propensity for accumulating somatic mutations varies along the genome, which critically influences somatic mosaicism, tumor evolution and the potential role of somatic mutations in the context of age-associated diseases. Genomic factors contributing to the variability of mutation rates have been established, including, for example, distance from the replication origin, chromatin structure and sequence context. However, their relative importance for explaining variable mutation rates along the genome as well as variable mutation rates between tissues remains elusive. Here, we present a modelling strategy that integrates 146 genomic features at different scales to predict susceptibilities for point mutations in 25 human tissues along the genome. These models faithfully predict mutation rates in coding and non-coding parts of the human genome in cancer and healthy tissues, including even unseen tissue types that were not used during the model training. Our work revealed that the dependency of mutation rates on chromatin structure and other genomic features is remarkably invariant across tissues, pointing to fundamental, conserved processes underlying mutagenic processes. Local variability in mutation rates on the scale of a few base pairs is almost exclusively driven by the sequence context, whereas large-scale variability is dominated by chromatin features, gene expression and GC content. Our modelling strategy quantifies the relative contribution of genomic factors to mutation susceptibility, predicting mutational biases at any genomic resolution across human tissues, and provides a basis for better understanding tumor evolution and age-related diseases.
-
10. Interpretable distillation reveals that deep learning splicing models suffer from pervasive confounders and blind spots.
PMID:日期:2026-08-21Predicting RNA splicing from genomic sequence is a crucial task for understanding gene regulation and interpreting genetic variation. Recent deep learning advancements have led to splicing prediction algorithms that achieve state-of-the-art performance compared to earlier models. However, owing to the limited interpretability of deep learning models, the predictive mechanisms of current splicing models remain poorly understood. Here we develop a framework to explain model prediction logic using interpretable distillation. Applying our framework, we find that RNA splicing prediction models suffer from pervasive confounders and blind spots, leading to poor performance on non-reference sequences. We find that splicing models recognize exons through surprisingly simple additive combinations of sequence motifs, including known splicing regulatory elements. Critically, our analysis also reveals that splicing models exploit genomic confounders unrelated to splicing and fail to adequately capture the effects of RNA structure, leading to systematic prediction errors. Our findings illuminate fundamental limitations of training models on genomic sequences and suggest ways to overcome them.