Hepatitis B virus (HBV) and hepatitis C virus (HCV) are unrelated non-cytopathic viruses that infect the liver. Hundreds of millions of people worldwide are chronic carriers of HBV or HCV, and together these viral infections are responsible for ~80% of hepatocellular carcinoma (HCC). Numerous reports suggest that certain microRNAs (miRNAs, small non-coding RNAs that regulate gene expression post-transcriptionally) are important in the pathogenesis of viral hepatitis, although very few studies have assessed them comprehensively. Furthermore, no study of viral hepatitis or HCC has analyzed other classes of small RNAs, such as tRNA-derived RNAs (tDRs). tDRs have recently been suggested to play a role in human health, although they have never been previously assessed in primary human tissue. Our goal was to characterize thoroughly the small RNA transcriptome in chronic viral hepatitis and HCC. We used high-throughput sequencing to conduct an unbiased profiling of small (14-40 nts) RNAs in liver from Japanese subjects with advanced hepatitis B or C and paired HCC. We also analyzed gene expression data (microarray) and determined that pathways controlling cholesterol homeostasis were among the most significantly altered in chronic viral hepatitis and especially in HCC. An integrative analysis of miRNA and gene expression data predicted several candidate master miRNA regulators of each disease state, including miR-21 and miR-27. We validated in cell culture the roles of miR-21 and miR-27 as regulators of cholesterol synthesis, suggesting that these two miRNAs may be responsible in part for cholesterol imbalance in chronic viral hepatitis and HCC. We next sought to assess tDRs in chronic viral hepatitis and liver cancer. From the small RNA-seq, we observed that tDRs were abundant and significantly increased in chronic viral hepatitis. Remarkably, tDR abundance exceeded that of miRNAs in most infected non-cancerous tissue. In cancer tissue, the total abundance of tDRs was reduced and the relative abundance of individual tDRs was altered. Despite the potential relevance of tDRs to human health and disease, there is no standardized nomenclature, and no method has yet been developed to quantify tDRs from small RNA-seq datasets. To ameliorate these limitations, I developed a publically available method called tDRmapper.