Classification and Regression Analysis of Lung Tumors from Multi-level Gene Expression Data

Abstract

We study classification and regression problems in lung tumours where high throughput gene expression is measured at multiple levels: epi-genetics, trancription and protein. We uncover the correlates of smoking and gender-specificity in lung tumors. Different genes are indicative of smoking levels, gender and survival rates at these different levels. We also carry out an integrative anaysis, by feature selection from the pool of all three levels of features. Our results show that the epigenetic information in DNA methylation is a better marker for smoking status than gene expression either at the transcript or protein levels. Further, surprisingly, integrative anlysis using multi-level gene expression offers no significant advantage over the individual levels in the classification and survival prediction problems considered.

Identification of Novel Lung Cancer Driver Genes Connecting Different Omics Levels With a Heat Diffusion Algorithm

“…As for their respective contribution at epigenomics and genomics levels, WNT9B (Lan et al, 2006;Farkas et al, 2014), POU3F3 (Li et al, 2014;Kumar et al, 2016), and PAX7 (Starzyńska et al, 2020) have all been shown to be associated with lung cancer at epigenomics and genomics levels independently. For the two remaining genes, NLGN4Y (ENSP00000342535) and TBR1 (ENSP00000374205), in 2019, researchers from University of Southampton summarized NLGN4Y as a multi-omics level driver for lung cancer at least at epigenomics and transcriptomics levels (Jeyananthan and Niranjan, 2019). As for the genomics level, variants in NLGN4Y can regulate cell proliferation in multiple pathogenesis, though not directly reported in lung cancer (Nardello et al, 2021).…”