Abstract Genomic selection (GS) leverages genome-wide markers and phenotypes to predict breeding values, with its effectiveness largely dependent on the accuracy of genomic prediction (GP) models. However, GP methods often struggle to capture inter-individual variability and are limited by the curse of dimensionality, where the number of SNPs far exceeds the sample size. To address these challenges, we present CLCNet (Contrastive Learning and Chromosome-aware Network), a novel deep learning framework that integrates contrastive learning and chromosome-aware feature modeling. CLCNet comprises two key components: (i) a contrastive learning module that enhances the model’s ability to capture fine-grained, genotype-dependent phenotypic differences among individuals, and (ii) a chromosome-aware module that captures structured feature selection at both chromosome and genome levels, thereby distilling the most informative SNPs. We evaluated CLCNet across four crop species, covering ten agronomically important traits, and compared it with a diverse set of classical linear, machine learning, and deep learning models. CLCNet achieved superior prediction performance, with statistically significant improvements in Pearson correlation coefficient (PCC), ranging from 0.34% to 12.19% over baseline, together with reduced mean squared error (MSE). Performance gains were more pronounced for traits with moderate linkage disequilibrium (LD; r 2 = 0.21-0.36) and high heritability ( h 2 > 0.66), such as those in maize, rapeseed, and soybean. For cotton traits characterized by high LD ( r 2 = 0.74) and lower heritability ( h 2 < 0.50), CLCNet maintained robust performance without degradation. Overall, these results demonstrate that CLCNet is an effective framework for improving genomic prediction accuracy and holds strong potential for practical applications in plant breeding. Short abstract CLCNet is a novel deep learning framework for genomic prediction that integrates contrastive learning with chromosome-aware feature selection. By jointly modeling inter-individual genotype–phenotype variation and chromosomal genomic structure, CLCNet improves prediction accuracy under high-dimensional, low-sample-size conditions. Across four crop species and ten agronomic traits, CLCNet consistently outperformed classical statistical, machine learning, and existing deep learning models. The framework also identified biologically relevant SNPs and candidate genes, demonstrating its potential for practical applications in genomic selection and computational plant breeding. Key points We propose CLCNet, a multi-task deep learning framework that integrates contrastive learning with chromosome-aware feature selection for genomic prediction, under high-dimensional, low-sample-size conditions. The chromosome-aware module explicitly exploits genomic structural information to select representative and informative SNPs across chromosomes. Contrastive learning improves model robustness by stabilizing representation learning and reducing the influence of random effects across samples. By complementing GWAS analyses, CLCNet provides additional insights into genotype–phenotype relationships with potential relevance for gene discovery. Biographical Note Jiangwei Huang is a PhD candidate at the Institute of Genetics and Developmental Biology, Chinese Academy of Sciences. His research focuses on genomic prediction, deep learning, and computational plant breeding. Zhihan Yang is a PhD candidate at the Institute of Genetics and Developmental Biology, Chinese Academy of Sciences. Her research interests include genomic prediction and bioinformatics. Rongcheng Han is an associate professor at the Institute of Genetics and Developmental Biology, Chinese Academy of Sciences. His research focuses on bioinformatics and plant phenomics. Yuqiang Jiang is a professor at the Institute of Genetics and Developmental Biology, Chinese Academy of Sciences. His research interests include plant genomics, plant phenomics and genetic improvement. Organization description The Institute of Genetics and Developmental Biology, Chinese Academy of Sciences, is a leading research institute focusing on genetics, genomics, molecular breeding, bioinformatics, and systems biology in plants and animals.

Motivation: The complex cellular networks underlying phenotypes are formed by the interacting gene modules. Building and analyzing genome-wide and high-quality Gene Co-expression Networks (GCNs) is useful for uncovering these modules and understanding the phenotypes of an organism. Results: Using large-scale RNA-seq samples, we constructed high coverage and confident GCNs in two monocot species rice and maize, and two eudicot species Arabidopsis and soybean, and subdivided them into co-expressed gene modules. Taking rice as an example, we discovered many interesting and valuable modules, for instance, pollen-specific modules and starch biosynthesis module. We explored the regulatory mechanism of modules and revealed synergistic effects of gene expression regulation. In addition, we discovered that the modules conserved among plants participated in basic biological processes, whereas the species-specific modules were involved in spatiotemporal-specific processes linking genotypes to phenotypes. Our study suggests gene regulatory relationships and modules relating to cellular activities and agronomic traits in several model and crop plants, and thus providing a valuable data source for plant genetics research and breeding. Availability and implementation: The analyzed gene expression data, reconstructed GCNs, modules and detailed annotations can be freely downloaded from ftp://47.94.193.106/pub. Supplementary information: Supplementary data are available at Bioinformatics online.

BACKGROUND: Evolutionary analysis provides a formal framework for comparative analysis of genomic and other data. In evolutionary analysis, observed data are treated as the terminal states of characters that have evolved (via transitions between states) along the branches of a tree. The NEXUS standard of Maddison, et al. (1997; Syst. Biol. 46: 590-621) provides a portable, expressive and flexible text format for representing character-state data and trees. However, due to its complexity, NEXUS is not well supported by software and is not easily accessible to bioinformatics users and developers. RESULTS: Bio::NEXUS is an application programming interface (API) implemented in Perl, available from CPAN and SourceForge. The 22 Bio::NEXUS modules define 351 methods in 4229 lines of code, with 2706 lines of POD (Plain Old Documentation). Bio::NEXUS provides an object-oriented interface to reading, writing and manipulating the contents of NEXUS files. It closely follows the extensive explanation of the NEXUS format provided by Maddison et al., supplemented with a few extensions such as support for the NHX (New Hampshire Extended) tree format. CONCLUSION: In spite of some limitations owing to the complexity of NEXUS files and the lack of a formal grammar, NEXUS will continue to be useful for years to come. Bio::NEXUS provides a user-friendly API for NEXUS supplemented with an extensive set of methods for manipulations such as re-rooting trees and selecting subsets of data. Bio::NEXUS can be used as glue code for connecting existing software that uses NEXUS, or as a framework for new applications.