Core Viewpoint - The article discusses the development of ANNEVO, a novel genome annotation method that utilizes a mixed expert architecture to achieve high-accuracy de novo gene annotation without relying on external evidence such as RNA sequencing or homologous proteins [3][6]. Group 1: ANNEVO Methodology - ANNEVO addresses two key challenges: modeling evolutionary heterogeneity and long-distance context modeling [5]. - The model distinguishes between different biological taxa at a macro level to minimize signal interference from distantly related species, while internally it learns specific gene structure patterns through a mixed expert mechanism [5]. - It incorporates a long-distance context modeling module to adapt to the complex features of genomic sequences, balancing local and global patterns [5]. Group 2: Research Findings - The method demonstrates the ability to model evolutionary patterns across different biological groups and long-distance sequence dependencies, achieving high-precision de novo gene annotation solely from DNA sequences [6]. - ANNEVO shows excellent generalization capabilities across multiple phylogenetic branches and can correct errors in existing reference databases, providing a new technical pathway for genome annotation [6]. - The research indicates a shift from traditional methods that heavily rely on external experimental data to a more intelligent and automated approach in gene annotation [6]. Group 3: Implications and Future Directions - ANNEVO represents a significant breakthrough in gene annotation technology, enhancing China's independent innovation capabilities in key genomic technologies [6]. - The method is expected to play a role in broader genomic functional analysis scenarios, particularly as it expands to more complex annotation tasks like non-coding RNA and alternative splicing [6]. - The research aligns with national strategies for biological security and the integration of artificial intelligence with life sciences, marking a critical direction for advancing the frontiers of life sciences [8].