We study how skeletal stem and progenitor cells, chondrocytes and osteoblasts change during growth, ageing and joint disease, using single-cell RNA sequencing (scRNA-seq), spatial transcriptomics and 3D light-sheet imaging. The project has two computational parts. (1) Processing and integrating large scRNA-seq datasets: our own mouse growth-plate time course and public mouse and human bone, bone-marrow and joint datasets from GEO/ENA, in total several hundred samples. This involves read alignment and quantification (STARsolo/Cell Ranger), quality control, batch integration, pseudotime/lineage analysis, gene-regulatory network inference and repeated permutation-based statistics, which need far more memory and core-hours than are available on a workstation. (2) Structure-based triage of candidate ligand-receptor interactions: candidates are first prioritised from co-expression across the single-cell atlas, then a few hundred pairs are modelled with AlphaFold-Multimer/ColabFold and Boltz-2, which require large-memory GPUs. Results will feed publications on growth-plate cell fate and osteoporosis-relevant signalling.