GM) 9 10In 2018 alone, an estimated 20,000 new acute myeloid leukemia (AML) patients 11 were diagnosed, in the United States, and over 10,000 of them are expected to 12 die from the disease. AML is primarily diagnosed among the elderly (median 68 13 years old at diagnosis). Prognoses have significantly improved for younger 14 patients, but in patients older than 60 years old as much as 70% of patients will 15 die within a year of diagnosis. In this study, we conducted stratified 16 computational meta-analysis of 2,213 acute myeloid leukemia patients compared 17 to 548 healthy individuals, using curated publicly available data. We carried out 18 analysis of variance of normalized batch corrected data, including considerations 19 for disease, age, tissue and sex. We identified 974 differentially expressed probe 20 sets and 4 significant pathways associated with AML. Additionally, we identified 21 70 sex-and 375 age-related probe set expression signatures relevant to AML. 22Finally, we used a machine learning model (KNN model) to classify AML patients 23 2 compared to healthy individuals with 90+% achieved accuracy. Overall our 24 findings provide a new reanalysis of public datasets, that enabled the 25 identification of potential new gene sets relevant to AML that can potentially be 26 used in future experiments and possible stratified disease diagnostics. 27 28 29 30 classification is highly dependent on the presence of chromosomal abnormalities, 47including chromosomal deletions, duplications, translocations, inversions, and 48 gene fusion. Mostly, AML is diagnosed through microscopic, cytogenetics, and 49 molecular genetic analyses of patients' blood and/or bone marrow samples. 50Microscopic examination is used to detect distinctive features (e.g. Auer rods) in 51 cell morphology, cytogenetic analysis to identify chromosomal structural 52 aberrations (e.g., t(8;21), inv(16), t(16;16), or t(9;11)), and molecular genetic 53 analysis to identify gene fusion (e.g., RUNX1-RUNX1T1 and CBFB-MYH11), and 54 mutations in genes frequently mutated in AML (e.g., NPM1, CEBPA, RUNX1, 55 FLT3) 6-8 . These cytogenetic and molecular genetic analyses are used to identify 56 prognosis markers that can be used to classify AML patients into three risk 57 categories: favorable, intermediate, and unfavorable. The largest group of AML 58 patients (almost 50%) however, present normal karyotype and lack genetic 59 abnormalities 7-10 . These patients are classified as intermediate risk, and often 60 have heterogeneous clinical outcome with standard therapy with risk of AML 61 relapse 11 . 62 63Additionally, AML prognosis worsens as age increases, and older patients 64 respond less to current treatments with poorer clinical outcomes than their 65 younger counterparts 12,13 . AML can occur in people of all ages but is primarily 66 diagnosed among the elderly (>60 years), with a median age of 68 year at 67 diagnosis 4 . Recent advances in AML biology expanded our understanding of its 68 complex genetic landscape and led to significant improv...