Atrial fibrillation (AF) is the commonest arrhythmia requiring intervention; we report here analyses of 2 ongoing studies of AF from the Pharmacogenomics of Arrhythmia Therapy (PAT) group. (1) ACE polymorphism and drug response: The Angiotensin Converting Enzyme (ACE) DD genotype increases plasma ACE, is presumed to thereby increase tissue angiotensin II, and has been linked to adverse outcomes in cardiovascular disease. Increasing evidence links activation of the renin-angiotensin system to the pathophysiology of atrial fibrillation (AF), suggesting the hypothesis that ACE I/D genotype modulates response of AF to therapy with conventional antiarrhythmic drugs (AADs). We have tested this idea in 229 patients (159 men, 70 women; age 51±15 years) prospectively enrolled in the Vanderbilt AF Registry. AAD outcome was defined blind to genotype as response if >75% reduction in frequency/duration of AF or non-response if duration/frequency was unchanged, necessitating a change in drug/ therapy. The frequencies of the DD, ID and II genotypes were in Hardy-Weinberg equilibrium (26%, 45% and 28% respectively). 42% of DD subjects were non-responders, compared to 25% of ID and 11% of II (P<0.01, Fisher's exact test). ID/DD patients had 1.5±0.7 AAD trials, compared to 1.3±0.6 in II subjects (P=0.03). (2) Post-operative AF: Clinical and genomic predictors of post-operative AF were studied in 940 elective cardiac surgery patients; 250 had AF, and 250 were selected as controls. Multifactor Dimensionality Reduction (MDR) analysis was used to determine the roles of clinical risk factors (age, gender, type of operation, preoperative medications, medical history, and postoperative events) and the candidate genetic markers interleukin-6 (IL-6) -174 C/G, ACE I/D, and apolipoprotein E (ApoE) alleles. History of AF, length of stay, age, duration of cardiopulmonary bypass, and beta-blocker therapy on discharge predicted AF. We also detected a single locus effect of the IL-6 -174 C/G polymorphism, which was able to predict post-operative AF with 59.8% accuracy. ACE I/D polymorphism and ApoE alleles did not show consistent effects in predicting AF, either alone or interacting with environmental variables. Taken together, these results highlight the likelihood that this common arrhythmia may have multiple etiologies that are in part genetically-determined, and demonstrate utility of novel computational approaches in assessing candidate susceptibility genes. Clinical Pharmacology & Therapeutics (2005) 77, P7–P7; doi: 10.1016/j.clpt.2004.11.029