Comprehensive multi-mark and multi-omic profiling reveals epigenetic and transcriptional reprogramming in lung tumors

We present comprehensive epigenomic and transcriptomic mapping of 18 tumor and 20 non-neoplastic tissues from non-small cell lung adenocarcinoma patients. Our profiling covers active marks (H3K4me3, H3K4me1, and H3K27ac), repressive marks (H3K27me3 and H3K9me3) and gene expression using only 20 mg of tissue. Genome-wide differentially modified peaks and differentially expressed genes associated with non-small cell lung cancer were identified. Key pathways related to cancer and cell proliferation were found to be enriched among these regions. 118 differential transcription factors (TF) were identified by Taiji, showing a global impact on TF in tumor. Regulation associated modules (RAM) analysis by findRAM uncovered consensus RAMs (cRAM) specific to non-plastic or tumor samples, most of which were marked with differential histone modification and gene expression. Integrative analysis by EpiSig identified 6 sections of genomic regions with distinct co-modification patterns. These clusters showed a clear distribution pattern over the specific cRAMs and were enriched in key cancer related pathways. Our results demonstrate the power of integrated analysis of multiple epigenomic and transcriptomic marks in patient samples. Chromatin immunoprecipitation followed by sequencing using a microfluidic platform (MOWChIP-seq) for histone modifications H3K27ac, H3K4me1, H3K4me3, H3K27me3 and H3K9me3, and RNA-seq using Smart-Seq2 in patient derived non-neoplastic lung tissue and non-small cell lung adenocarcinoma samples. Please note that the raw ChIP-seq and RNA-seq data on the human lung tissues were deposited in dbGaP under accession number phs003113.v1.p1