Additional file 2: Supplement Table 1. Lists of genes included within each gene expression signature analyzed to determine Active/Inactive, cAMP lipolysis, adipocyte browning, SASP, AST, IGF1, IGF1R, IFN, TGFβ, and CSR activity scores. Also indicated is their reference sources and method of score calculation.

Additional file 2: Supplement Table 1. Lists of genes included within each gene expression signature analyzed to determine Active/Inactive, cAMP lipolysis, adipocyte browning, SASP, AST, IGF1, IGF1R, IFN, TGFβ, and CSR activity scores. Also indicated is their reference sources and method of score calculation.

Additional file 3: Supplement Table 2. RNAseq cancer gene hotspot mutations detected in 151 KTB samples. Sheet 1 identifies the > 5100 unthresholded cancer mutation hotspot calls from the KTB RNAseq analysis occurring in the combined set of MSK-IMPACT curated cancer hotspot clinical targets [10] and the experimentally determined RNAseq identified set of expressed cancer gene hotspot mutations within > 6700 normal GTEx human tissue samples as recently described [11]. Sheet 2 (“normal_breast_rna.getz_list.092”) lists only those > 1760 cancer gene hotspot mutations from sheet 1 meeting the threshold mutation likelihood score > 5, while sheets 3 and 4 list the thresholded hotspot mutations according to F and P sample batches with each sample phenotyped as either Active or Inactive. Sheet 5 is a sample key mapping all KTB identification numbers, barcodes, and UUID numbers pertaining to the RNAseq results and cancer gene hotspot mutation calls. Of note, Fig. 2 panels (“Breast sample scatterplots and TumorMaps of RNA expressed cancer hotspot mutations”) include plots derived from the curated and thresholded hotspot mutations as listed in sheet 2.

Additional file 3: Supplement Table 2. RNAseq cancer gene hotspot mutations detected in 151 KTB samples. Sheet 1 identifies the > 5100 unthresholded cancer mutation hotspot calls from the KTB RNAseq analysis occurring in the combined set of MSK-IMPACT curated cancer hotspot clinical targets [10] and the experimentally determined RNAseq identified set of expressed cancer gene hotspot mutations within > 6700 normal GTEx human tissue samples as recently described [11]. Sheet 2 (“normal_breast_rna.getz_list.092”) lists only those > 1760 cancer gene hotspot mutations from sheet 1 meeting the threshold mutation likelihood score > 5, while sheets 3 and 4 list the thresholded hotspot mutations according to F and P sample batches with each sample phenotyped as either Active or Inactive. Sheet 5 is a sample key mapping all KTB identification numbers, barcodes, and UUID numbers pertaining to the RNAseq results and cancer gene hotspot mutation calls. Of note, Fig. 2 panels (“Breast sample scatterplots and TumorMaps of RNA expressed cancer hotspot mutations”) include plots derived from the curated and thresholded hotspot mutations as listed in sheet 2.

Additional file 4: Supplement Table 3. Rank ordered GSEA ( www.gsea-msigdb.org/gsea ) analysis showing 186 gene sets upregulated (from total set of 18,408 gene sets) in the batch-integrated Active vs. Inactive transcriptome samples, at FDR ≤ 10%. Also shown are their individual nominal p-values, FDR q-values, their gene set size, and the 21 (11.2%) that are specifically involved in adipose-associated pathways.

Additional file 5: Supplement Table 4. Master spreadsheet listing showing sample and donor covariates for each of the 151 KTB barcodes including batch assignment (F or P), Active/Inactive phenotype assignment and score, donor features (including age, BMI, 5 year Gail risk scores), percentages of adipocyte/stromal/epithelial cell nuclei, TDLU counts, mean adipocyte areas, immune modules, and all gene signatures and single genes values used to calculate the Pearson correlations shown in Fig. 5.

Additional file 4: Supplement Table 3. Rank ordered GSEA ( www.gsea-msigdb.org/gsea ) analysis showing 186 gene sets upregulated (from total set of 18,408 gene sets) in the batch-integrated Active vs. Inactive transcriptome samples, at FDR ≤ 10%. Also shown are their individual nominal p-values, FDR q-values, their gene set size, and the 21 (11.2%) that are specifically involved in adipose-associated pathways.

Additional file 5: Supplement Table 4. Master spreadsheet listing showing sample and donor covariates for each of the 151 KTB barcodes including batch assignment (F or P), Active/Inactive phenotype assignment and score, donor features (including age, BMI, 5 year Gail risk scores), percentages of adipocyte/stromal/epithelial cell nuclei, TDLU counts, mean adipocyte areas, immune modules, and all gene signatures and single genes values used to calculate the Pearson correlations shown in Fig. 5.