Beijing Genomics Institute

Lineages undergoing rapid evolutionary diversification may evolve a set of mixed traits as a result of hybrid speciation, gene flow, or incomplete lineage sorting (ILS). However, how best to quantify these alternative processes impacting phenotypic variation remains unclear. Here, we examined trait evolution in two sister genera of endangered Asian primates, Trachypithecus and Semnopithecus . By assembling a de novo genome of the Trachypithecus pileatus group, which is geographically located in a transitional zone between these two genera, our integrated phylogenomic analyses clarified the evolutionary relationship of the T. pileatus group with other Trachypithecus species. We also identified that morphological similarities shared between this species group and Semnopithecus are the result of ILS, rather than gene flow or hybrid speciation. Across all chromosomes of the T. pileatus group, ILS contributed 8.9% of whole genome segments. Across these segments, we distinguished 77 genes such as FGFBP1 and FOXO1 that are involved in pathways of bone development and osteoblast differentiation. Functional experiments indicate that FGFBP1 genotypes shared by species of the T. pileatus group and Semnopithecus appear to enhance osteogenic capability and mineralization, possibly resulting in larger body size and similarities in skull morphology compared with other species of Trachypithecus . The study reveals that ILS has shaped the evolution of mixed traits within gene-regulatory networks, driving phenotypic diversity during periods of rapid species divergence in this highly successful primate radiation. ILS appears to be more common than previously thought and represents a critical step in accurately assessing the phylogeny of closely related taxa.

Epigenome-wide association studies (EWAS) has become an indispensable approach for elucidating the epigenetic basis of complex traits. EWAS Open Platform (https://ngdc.cncb.ac.cn/ewas/) includes three main components: EWAS Atlas (curated associations from publications), EWAS Data Hub (normalized DNA methylation array data), and EWAS Toolkit (one-stop analysis services). Here, we present a new release of EWAS Open Platform with the following significant updates and enhancements: (i) Expanded information: EWAS Atlas houses over 800 000 associations and incorporates an additional 17 000 curated causal relationships. EWAS Data Hub contains >180 000 batch-corrected samples, which incorporate data from newly added Illumina MethylationEPIC v2.0 BeadChip (935K) as well as trait-specific methylation profiles. EWAS Toolkit now provides an online batch correction tool and an interactive epigenetic causal network. (ii) Enhanced interoperability: data, knowledge, and toolkit are fully interconnected through a unified retrieval, offering integrated summaries and visualization capabilities. (iii) Artificial intelligence (AI)-based service: the platform is newly equipped with an AI-assisted question-answering service, allowing users to interactively explore EWAS-related questions and generate tailored insights. Taken together, EWAS Open Platform has undergone a significant upgrade across data resources, analytical tools, and service functionalities, offering more advanced support for unraveling complex molecular mechanisms from an epigenomic perspective.

LncExpDB is a comprehensive expression database of human long non-coding RNAs (lncRNAs) across diverse biological contexts. Here, we present LncExpDB 2.0 (https://ngdc.cncb.ac.cn/lncexpdb/) with significant updates and enhancements as detailed below: (i) expansion of biological contexts from 9 to 15 by integrating immunotherapy, aging, metabolic diseases, neurodegenerative diseases, reproduction, and wound healing; (ii) identification of 44 752 highly expressed genes, 39 253 featured genes with function-linked expression patterns, and 35.19 million lncRNA–mRNA co-expression pairs; (iii) in-depth functional analyses of featured genes, such as correlation with tumor microenvironment, age-related expression profiles, and functional enrichment; (iv) deployment of new tools for exploring complex lncRNA–mRNA interactions and lncRNA–immune checkpoint correlation; and (v) development of a pipeline module with documented workflow and command-line instructions for reproducible expression profiling. Collectively, LncExpDB 2.0 provides a multi-context landscape of lncRNA expression and serves as a critical resource for functional exploration of the “dark matter” in the human genome.