Sequencing platform modeling and analysis.

Abstract

As the cost rapidly declines, the next-generation sequencing (NGS) technologies are becoming the main work horse of modern molecular biology to measure the abundance of genomic markers. NGS offers a number of technical advantages such as a greater dynamic range for measuring abundance and a higher resolution for low abundance molecules; moreover, it allows studying innovative biological problems such as the detection of novel molecular features and performing a more seamless integrative analysis of multiple types of molecular data for the same set of samples. Their utilization in cancer genomic studies such as The Cancer Genome Atlas (TCGA) presents unprecedented opportunities to advance our understanding of cancer. While certain aspects of NGS data analysis fall under the general framework of high-throughput data analysis and can be addressed by existing statistical methodologies, it also presents unique analytic challenges that require development of new statistical approaches. We take the opportunity of this special issue to highlight several recent methodological develop ments on NGS data analysis by leading researchers in the field. Some of the articles in this special issue are summarized below. • RNA Expression Level: Identification of molecular signatures is an important aspect of the analysis of RNA expression data. Ha et al developed a novel method in a causal structure learning framework to discover prognostic gene signatures. Their method represented the causal structure by directed acyclic graphs and constructed genespecific network modules to constitute a gene and its corresponding regulators; the method then correlated each module with a survival outcome to allow for a network oriented approach to select prognostic genes. They applied the new method to a clear cell renal cell carcinoma study from TCGA and found several novel prognostic genes. • RNA Splicing Variant: Alternative splicing is a posttranscriptional process that allows a single gene to produce multiple mRNA isoforms (namely, mRNA slicing variants). This process can be regulated by genetic elements such as single nucleotide polymorphisms (SNPs), which are called splicing quantitative trait loci (sQTL). The main analytic challenge of identifying sQTL is the estimation of isoform-specific mRNA expression based on RNA sequencing data. Jia et al evaluated three statistical Supplement Aims and Scope