Live attenuated vaccines against diseases caused by RNA viruses, such as polio, measles, mumps, and yellow fever, are among the most efficient and safest human vaccines. They induce lifelong protective immunity after 1 or 2 administrations in 90%–95% of immunized individuals and are easily manufactured at large scale and low cost. Despite their amazing clinical success in the global control of these diseases and despite the very low number of adverse events observed over decades of use in billions of individuals, these vaccines still raise worry of possible reversion to the original pathogenicity. Live attenuated vaccines have all been empirically derived from pathogenic clinical isolates by serial cultivation in vitro, and our understanding of the molecular basis of their attenuation is still limited. In this issue of the Journal, Beck et al [1] report results of deep sequencing analysis of the genome in of the yellow fever 17D (YF-17D) vaccine strain and its comparison to the parental strain Asibi from which it was originally derived in 1937 [2]. This first comparison of a live RNA virus vaccine to its wild-type parental strain by deep sequencing provides biologically significant observations on attenuation and critical information on the stability of live vaccines and the risk of reversion to virulence. Yellow fever is a viral hemorrhagic fever endemic in tropical and subtropical regions of sub-Saharan Africa and South America caused by the mosquito-borne yellow fever flavivirus (YFV). In the African rainforest, the zoonotic transmission cycle of YFV in great apes involves Aedes species mosquitoes, whereas the urban cycle in humans occurs through Aedes aegypti. The World HealthOrganization (WHO)estimates that there are 200 000 cases of and 30 000 deaths due to yellow fever annually. YFV infection may progress to multiorgan failure of the liver, kidneys, and myocardial tissues, resulting in hemorrhagic shock with a case-fatality varying from 20% to 50%. The infectious origin of yellow fever was proven by Walter Reed in 1901, and the virus was isolated in 1927 by Adrian Stokes from a Ghanaian patient named Asibi [3]. A few years later, Max Theiler managed to attenuate this virus while maintaining its immunogenicity, paving the way to the vaccine. He was awarded the Nobel Prize in Medicine in 1951 for this work. The YF-17D vaccine was empirically attenuated after 204 serial passages of the Asibi YFV strain in mouse and chicken embryo tissues. Variants were assessed for their ability to raise neutralizing antibodies and to protect monkeys from a challenge with the virulent Asibi strain. In 1945, the WHO granted the use of 2 substrain seed lots of this vaccine in humans: 17DD, which was used in South America, and 17D204, which is currently used in the rest of the world. Both substrains in the vaccine are manufactured in embryonic chicken eggs. Currently, 6 manufacturers produce YF-17D vaccine, using a seed lot system that has not changed since 1945 to ensure genetic stability and safety of the vaccine seeds. Another live yellow fever vaccine was developed in France in 1934 by attenuation in mouse brain tissue, and the French neurotropic vaccine (FNV) was widely used in Africa until 1982, when its production was stopped as a result of a high incidence of neurological complications. Vaccine seeds are now assessed for neurotropism and viscerotropism in primates before the production process is finalized. A single administration of YF17D vaccine induces long-lasting protection; however, the WHO recommends booster immunizations every 10 years. YF17D has an excellent safety record over years of massive use. Yet, rare adverse events are identified, occurring between 2 and 30 days after vaccination, comprising either neurotropic or viscerotropic diseases with a >50% fatality rate (0.05–1.5 cases per 100 000 vaccinations) [4]. The mechanisms underlying these rare complications remain largely unknown. Molecular and animal studies performed to date provide no evidence that mutations Received and accepted 4 October 2013. Correspondence: Frederic Tangy, PhD, Unite de Genomique Virale et Vaccination, CNRS, UMR 3569, Institut Pasteur, 28 rue du Dr Roux, Paris 75015, France (ftangy@pasteur.fr). The Journal of Infectious Diseases © The Author 2013. Published by Oxford University Press on behalf of the Infectious Diseases Society of America. All rights reserved. For Permissions, please e-mail: journals. permissions@oup.com. DOI: 10.1093/infdis/jit551