Acute myeloid leukemia is a rapidly progressive neoplasm that usually results in the demise of affected dogs within days to weeks. We currently know very little about the genetic basis of AML in dogs, which recapitulates the disease in humans. However, in humans, tremendous advances have been made in the understanding of the genetic basis of AML through the application of high-throughput DNA and RNA sequencing (RNA-seq). Genetic analysis now is the foundation on which this heterogeneous neoplasm is classified and is a requisite component of diagnosing the leukemia in human patients. Classification of AML into subtypes is of clinical relevance with prognostic and therapeutic implications in humans. For example, certain gene defects have a poorer outcome or respond to treatment with specific drugs. In contrast, we still use an old phenotypic classification scheme in dogs, which has little prognostic or therapeutic relevance, and we continue to treat dogs with the same drugs, even when they are of limited efficacy. Thus, it is vital that we learn more about the gene abnormalities and gene expression patterns of AML in dogs so we can better diagnose and treat the leukemia and provide owners with more accurate prognoses. We have an ongoing Canine Health Centerfunded study to perform whole exome sequencing (WES) in dogs with well-characterized AML. To date, we have isolated DNA from 38 dogs with AML and found multiple chromosome abnormalities in a few dogs on preliminary analysis. We have also isolated RNA from 25 of these 38 cases. Our goal with this proposal is to perform RNA-seq on these samples. Whole exome sequencing only identifies abnormalities in coding regions of the gene and misses important gene fusions (which comprise one of the major categories of human AML). In contrast, RNA-seq detects gene fusions and splice variants arising from intron defects. In addition, RNA-seq yields the functional consequences of abnormal genes on cellular pathways and metabolism, which provides information on treatment and prognosis, data not provided from WES. In Aim 1, we will test the hypothesis that dogs with AML have disease-associated gene fusions and splice variants that allow classification into different types by performing mutational analysis on mRNA. In Aim 2, we will perform gene expression profiling to test the hypothesis that the transcriptome will distinguish AML from acute lymphoid leukemia, and acute monocytic from myelomonocytic variants of AML. This study will be the first comprehensive analysis of the genome and transcriptome of dogs with AML and will provide a solid foundational database of the genetic basis of AML and potentially a new clinically relevant classification scheme in dogs.