Coexpression
TKU uniform-model v4 TPM and independent OH-1004 datasets are connected. Pearson candidates are computed within each study on log2(TPM+1), without pooling studies. Two-sample studies are excluded; legacy TK1151 caches remain quarantined.
Evidence boundary: the all-dataset view does not merge different projects or reference versions into one artificial matrix. Pearson correlations on log2(value+1) are computed within each RNA-seq dataset and then summarized. Results indicate expression similarity only, not regulation, interaction, or causality.
Data Summary
All datasetsThis page summarizes only the coexpression results in the available catalog. Studies are computed separately without mixing samples across datasets. Correlations from only 3 samples are especially unstable and should not be treated as high-confidence regulatory evidence. TKU uses per-gene, per-study top-200 query caches; this is neither an all-pairs network nor experimental regulatory evidence.