MicroRNAs (miRNA) are a class of small regulatory RNAs that mediate post-transcriptional silencing of specific target mRNAs. Our overall hypotheses are that miRNA expression is unique to tumor molecular phenotype; that miRNA expression levels at time of diagnosis predicts survival; and that miRNA expression is associated with inflammation-related genetic and lifestyle factors key to colorectal cancer (CRC). This study takes a two pronged approach to addressing our hypotheses. While we propose to validate previously identified miRNAs that have been identified as associated with CRC (either by differential expression or from assessment of mutations), we will add to the field through discovery of new and important associations that may be unique to specific molecular phenotypes, to polyp to cancer progression, and to survival. We will analyze the expression of 866 human miRNAs using data derived from tumor and paired normal tissue at time of diagnosis from: 1660 people with incident colon cancer; 840 people with incident rectal cancer; and 350 polyps from our colon and rectal cases who reported a prior polyp; 5% of tumors, will be analyzed for quality control. Our total assessment will be on 5475 samples. We will extend the validation of previously identified mutated miRNAs and differentially expressed miRNAs to determine if these alterations are associated with specific tumor molecular phenotype, inflammation-related factors, clinical factors and survival. Important miRNAs will be validated using TaqMan-based assays. Associations will be tested based on differential expression for both individual and groups of miRNAs using recent extensions of several statistical methods including ANOVA, logistic regression, and Cox proportional hazards models. Our sample size allows for both a training and validation component, and provides sufficient statistical power to meet the study goals. MiRNAs that are differentially expressed in polyps and in subsequent tumors will provide new insights into targets for screening and treatment and differential miRNAs that function as the "driver" vs. the "passenger" in the carcinogenic process. Testing of mutated miRNAs identified from sequencing in conjunction with tumor phenotype, clinical, and survival data will further validate the importance of these miRNAs, and provide insight as to which CRC molecular pathway the miRNAs function. Our rich dataset of lifestyle, genetic, clinical and prognosis, and tumor molecular phenotype on 2500 CRC and paired normal tissue allows us to examine factors that are associated with miRNAs in a large set of population-based cases. The miRNAs identified in these analyses will elucidate pathways important in the etiology of CRC and will provide insight into potential targets for screening and treatment.