Statistical Modeling for Genome Data Analysis to Detect Agricultural Biomarkers

Abstract

The focuses of this study were to evaluate the performance of different statistical methods from the perspective of various genomic data such as phenotypic-genotypic data, gene expression (microarray/RNA-Seq) data, SNP data and meta-genomic data collected from different environmental samples. We also performed some in silico analysis of RNA silencing machinery genes in wheat (Triticum aestivum) based on the RNAi genes of arabidopsis thaliana and expression profile analysis of seven TaDCL genes in leaves and roots as well as against drought stress using qRT-PCR. In Chapter Two, we explored better QTL mapping approach by comparative study. We found that Composite Interval Mapping (CIM) performs significantly better than the other four Simple Interval Mapping (SIM) methods in detecting QTL positions in backcross technique both on simulated data and on real rice genome dataset. In the case of real rice genome data analysis for backcross population, the CIM identified some vital positions that were not detected by the traditional SIM approaches.-----