One of the big aims in human genetics is to understand the biological mechanism underlying the genetic associations. In the past decades, the rapid development of biotechnology has made tremendous progress to approach this aim. For instance, with advanced and specialized devices and data automation systems, more complex phenotypes can be measured at higher accuracy and in more individuals. And with the inventions in high-throughput sequencing, we can profile various types of biological molecules in organs, tissues, and cells. As a geneticist, we face a massive amount of biological data at different levels and of great diversity, creating unprecedented opportunities for making discoveries. However, making the best use of data and translating them into scientific insights remain challenging. In the current data-dominated era, statistical modeling has become a vital tool to fill the gap between biological data and scientific discoveries. My dissertation spans multiple topics in statistical genetics involving the handling of genomic, transcriptomic, and phenomic data. In Chapter 2, I pro- pose a unified statistical framework, along with computationally efficient implementation, leveraging signals from both total counts and allele-specific counts to study the genetic effect of variants on cis-regulation. In Chapter 3, I show the utility of predicted transcriptome-based polygenic risk scores in terms of the prediction performance in the matched ancestry and cross ancestry. In Chapter 4, I propose a method to impute the parental origin of the haplotypes by exploiting the parental phenome and analyze the potential benefit of using these imputed haplotypes in a GWAS with parental phenotypes and offspring genotypes. In Chapter 5, I design and implement a data analysis pipeline studying the relation between magnetic resonance imaging-derived brain features and complex phenotypes by leveraging genetic evidence rather than purely observational data. Besides methodological advancements, I also involve in collaborative efforts on analyzing and integrating the state-of-the-art datasets to decipher the genetic basis of transcriptome in multi-tissue setting and how it relates to complex phenotype genetics, which is shown in Appendix A.