Top2All: Benchmarking sparse-to-transcriptome reconstruction from high-expression profiles across sex- and sex-steroid-defined conditions

Reproducibility archive accompanying the manuscript “Top2All: Benchmarking sparse-to-transcriptome reconstruction from high-expression profiles across sex- and sex-steroid-defined conditions.” This archive contains the exact curated 24-sample expression inputs, ranked sparse TopN inputs, formal Top2All prediction files used for evaluation, detailed numerical evaluation outputs, sample and validation-fold metadata, gene-order information, provenance manifests, and SHA-256 checksums. All included project-origin assets were verified against pre-existing production hashes recovered from the Kyushu University Genkai computing environment. Raw RNA-seq data are available separately from the DNA Data Bank of Japan (DDBJ) under accession DRA015952. Project-trained checkpoints (~23.9 GB), pretrained Mouse-Geneformer resources, token dictionaries, raw sequencing reads, fonts, and other third-party resources are not included in this core archive. The associated source-code repository is publicly available at https://github.com/takbaba-lab/Top2All-Reproducibility.