Statistical methods for integrating multi-omics sequence data and unveiling molecular underpinnings of non-small cell lung cancer

The past decade has witnessed tremendous advancements in sequencing technologies, enabling collection of highly accurate and high resolution data on the structure and function of the genome. This opens great opportunities to tackle many fundamental questions in medicine, agriculture and environmental biology.

Funded by an international collaborative project, we have collected genomic DNA, mRNA, microRNA and MeDIP (Methylated DNA Immunoprecipition) sequence data from about 200 carefully scrutinized pairs of NSCLC (Non-small cell lung cancer) tissue and the corresponding paratumorous tissue samples, as well as from 20 cell lines of 3 major NSCLC pathological types. Based on these sequencing datasets, the project is proposed to develop statistical methods and computational tools for integrating the multilayers of omic sequence data to identify the molecules and the interactions which have significant influence on progression, development and metastasis of NSCLC. The project will involve case and control based genome wide association analysis (GWAS) with genomic DNA, RNA, and microRNA expression data, analysis of multi-dimensional network construction, causal relationship prediction, functional module diagnosis from the expression and regulation network etc.

Although the project uses sequencing datasets from lung cancer samples, it presents a generic bioinformatics and statistical question of how different omics sequencing data can be integrated to differentiate biological case and control treatments. Thus, focus of the project will be on development of efficient statistical algorithms and the corresponding bioinformatic tools for modelling and analysing biological datasets of the similar kind. It has thus significant implications to genomic analyses with plant and animal species.