Integrative Analysis for Multi-Omics Data in Non-Small-Cell Lung Cancer

Moores Cancer Center, UC San Diego. All tumor samples were in pathological stages I and II. 8 pairs of samples were "matched", meaning that the tumor and normal tissues came from the same patient. Microfluidic oscillatory washing–based chromatin immunoprecipitation followed by sequencing (MOWChIP-seq) was used to profile the binding of H3K4me1, H3K4me3, H3K9me3, H3K27ac and H3K27me3. SmartSeq-2 was used to map mRNA expression. We identified a large number of differentially modified regions for both epigenetic and transcriptomic markers between tumors and normal tissues. Deep learning model showed extensive transcription factor network rewiring. Raw sequencing data of the ChIP-seq and RNA-seq libraries are available through this dbGaP submission.]]>
Samples were collected from deidentified human subjects and stored by Biorepository and Tissue Technology Shared Resource at Moores Cancer Center, UC San Diego. We requested for all samples that can provide 50 mg of tissue. All tumor samples were of the pathological stages I and II. One stage III tumor sample was provided by the tissue bank but did not yield libraries of sufficient enrichment for sequencing.]]>