S1 Text - Machine learning on multiple epigenetic features reveals H3K27Ac as a driver of gene expression prediction across patients with glioblastoma

Section S1. Epigenetic marker and RNA-sequencing information. Section S2. Flowcharts of data preparation and preprocessing process. Section S3 . Supporting predictive models’ information [ 36 – 38 ]. Section S4 . Correlation highlights the variation among the epigenetic features for each dataset. Section S5. Supporting model results. Section S6. Computational considerations. Section S7. Perturbation results support model feature importances. Section S8. Supporting gene expression count and model error analysis. Section S9. Single-patient experimental setup demonstrates comparable model prediction performance to cross-patient modeling and echoes the importance of H3K27Ac. Section S10. Model evaluation over subsets of data supports the trends uncovered in the results. (DOCX)