BMI trajectory of 8,155,894 Japanese adults from exhaustive health checkup data: the contributions of age-related changes in height and weight
Related Articles
Evolutionary optimization of model merging recipes
Large language models (LLMs) have become increasingly capable, but their development often requires substantial computational resources. Although model merging has emerged as a cost-effective promising approach for creating new models by combining existing ones, it currently relies on human intuition and domain knowledge, limiting its potential. Here we propose an evolutionary approach that overcomes this limitation by automatically discovering effective combinations of diverse open-source models, harnessing their collective intelligence without requiring extensive additional training data or compute. Our approach operates in both parameter space and data flow space, allowing optimization beyond just the weights of the individual models. This approach even facilitates cross-domain merging, generating models such as a Japanese LLM with math reasoning capabilities. Surprisingly, our Japanese math LLM achieved state-of-the-art performance on a variety of established Japanese LLM benchmarks, even surpassing models with substantially more parameters, despite not being explicitly trained for such tasks. Furthermore, a culturally aware Japanese vision–language model generated through our approach demonstrates its effectiveness in describing Japanese culture-specific content, outperforming previous Japanese vision–language models. This work not only contributes new state-of-the-art models back to the open-source community but also introduces a new paradigm for automated model composition, paving the way for exploring alternative, efficient approaches to foundation model development.
Composite variable bias: causal analysis of weight outcomes
Researchers often use composite variables (e.g., BMI and change scores). By combining multiple variables (e.g., height and weight or follow-up weight and baseline weight) into a single variable it becomes challenging to untangle the causal roles of each component variable. Composite variable bias—an issue previously identified for exposure variables that may yield misleading causal inferences—is illustrated as a similar concern for composite outcomes. We explain how this occurs for composite weight outcomes: BMI, ‘weight change’, their combination ‘BMI change’, and variations involving relative change.
Timing based clustering of childhood BMI trajectories reveals differential maturational patterns; Study in the Northern Finland Birth Cohorts 1966 and 1986
Children’s biological age does not always correspond to their chronological age. In the case of BMI trajectories, this can appear as phase variation, which can be seen as shift, stretch, or shrinking between trajectories. With maturation thought of as a process moving towards the final state – adult BMI, we assessed whether children can be divided into latent groups reflecting similar maturational age of BMI. The groups were characterised by early factors and time-related features of the trajectories.
The genomic landscape of gene-level structural variations in Japanese and global soybean Glycine max cultivars
Japanese soybeans are traditionally bred to produce soy foods such as tofu, miso and boiled soybeans. Here, to investigate their distinctive genomic features, including genomic structural variations (SVs), we constructed 11 nanopore-based genome references for Japanese and other soybean lines. Our assembly-based comparative method, designated ‘Asm2sv’, identified gene-level SVs comprehensively, enabling pangenome analysis of 462 worldwide cultivars and varieties. Based on these, we identified selective sweeps between Japanese and US soybeans, one of which was the pod-shattering resistance gene PDH1. Genome-wide association studies further identified several quantitative trait loci that accounted for large-seed phenotypes of Japanese soybean lines, some of which were also close to regions of the selective sweeps, including PDH1. Notably, specific combinations of alleles, including SVs, were found to increase the seed size of some Japanese landraces. In addition to the differences in cultivation environments, distinct food processing usages might result in changes in Japanese soybean genomes.
Single-cell transcriptomic atlas of the human testis across the reproductive lifespan
Testicular aging is associated with declining reproductive health, but the molecular mechanisms are unclear. Here we generate a dataset of 214,369 single-cell transcriptomes from testicular cells of 35 individuals aged 21–69, offering a resource for studying testicular aging and physiology. Machine learning analysis reveals a stronger aging response in somatic cells compared to germ cells. Two waves of aging-related changes are identified: the first in peritubular cells of donors in their 30s, marked by increased basement membrane thickness, indicating a priming state for aging. In their 50s, testicular cells exhibit functional changes, including altered steroid metabolism in Leydig cells and immune responses in macrophages. Further analyses reveal the impact of body mass index on spermatogenic capacity as age progresses, particularly after age 45. Altogether, our findings illuminate molecular alterations during testis aging and their relationship with body mass index, providing a foundation for future research and offering potential diagnostic markers and therapeutic targets.
Responses