Meth3D-Net V6: Analysis Code and Processed Epigenomic Outputs for Multi-Scale Medulloblastoma Validation

This record contains analysis notebooks and processed output files for Meth3D-Net V6, an integrative multi-scale epigenomic framework for medulloblastoma. Included: seven Jupyter notebooks (NB01–NB07) implementing genome-wide methylation prediction, chromatin instability scoring, TCGA pan-cancer validation (n=9,854 samples, 33 cancer types), lncRNA candidate prioritisation, ceRNA network construction, and Epigenetic Senescence Susceptibility Index (ESSI) computation. Processed outputs include Layer A/B/C validation summaries, per-chromosome enrichment results, lncRNA priority scores, lncRNA–miRNA pairs, and 49 validated ceRNA triplets. Large probe-level matrices (105,451 probes × 9,854 samples) are hosted on Kaggle at https://www.kaggle.com/neetuaashi/tcga-pancan-methylation due to file size constraints. All primary input data are from publicly available repositories (GEO: GSE19418, GSE85212, GSE186599, GSE16256; TCGA PanCanAtlas). Associated manuscript: Singh N. Meth3D-Net V6: Multi-Scale Epigenomic Framework for Methylation, Chromatin Instability, and lncRNA Regulation in Medulloblastoma. [In preparation, 2026]