Dataset Information

SIMON, an Automated Machine Learning System, Reveals Immune Signatures of Influenza Vaccine Responses.

ABSTRACT: Machine learning holds considerable promise for understanding complex biological processes such as vaccine responses. Capturing interindividual variability is essential to increase the statistical power necessary for building more accurate predictive models. However, available approaches have difficulty coping with incomplete datasets which is often the case when combining studies. Additionally, there are hundreds of algorithms available and no simple way to find the optimal one. In this study, we developed Sequential Iterative Modeling "OverNight" (SIMON), an automated machine learning system that compares results from 128 different algorithms and is particularly suitable for datasets containing many missing values. We applied SIMON to data from five clinical studies of seasonal influenza vaccination. The results reveal previously unrecognized CD4+ and CD8+ T cell subsets strongly associated with a robust Ab response to influenza Ags. These results demonstrate that SIMON can greatly speed up the choice of analysis modalities. Hence, it is a highly useful approach for data-driven hypothesis generation from disparate clinical datasets. Our strategy could be used to gain biological insight from ever-expanding heterogeneous datasets that are publicly available.

SUBMITTER: Tomic A

PROVIDER: S-EPMC6643048 | biostudies-literature | 2019 Aug

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

SIMON, an Automated Machine Learning System, Reveals Immune Signatures of Influenza Vaccine Responses.

Tomic Adriana A Tomic Ivan I Rosenberg-Hasson Yael Y Dekker Cornelia L CL Maecker Holden T HT Davis Mark M MM

Journal of immunology (Baltimore, Md. : 1950) 20190614 3

Machine learning holds considerable promise for understanding complex biological processes such as vaccine responses. Capturing interindividual variability is essential to increase the statistical power necessary for building more accurate predictive models. However, available approaches have difficulty coping with incomplete datasets which is often the case when combining studies. Additionally, there are hundreds of algorithms available and no simple way to find the optimal one. In this study, ...[more]

PMID: 31201239

Similar Datasets

Project description:During the past decade, a number of H5 subtype influenza vaccines have been developed and tested in clinical trials, but most of them induced poor serum antibody responses prompting the evaluation of novel vaccination approaches. One of the most promising ones is a "prime-boost" strategy, which could result in the induction of prompt and robust immune responses to a booster influenza vaccine following priming with homologous or heterologous vaccine strains. In our study we evaluated immunogenicity of an adjuvanted A(H5N1) inactivated influenza vaccine (IIV) in healthy adult subjects who received A(H5N2) live attenuated influenza vaccine (LAIV) 1.5 years earlier and compared this with a group of naïve subjects. We found that priming with A(H5N2) LAIV induced a long-lasting B-cell immunological memory against influenza A(H5N1) virus, which was brought on by more prompt and vigorous antibody production to a single dose of A(H5N1) IIV in the primed group, compared to the naïve controls. Thus, by day 28 after the first booster dose, the hemagglutination inhibition and neutralizing (MN) antibody titer rises were 17.2 and 30.8 in the primed group, compared to 2.3 and 8.0 in the control group, respectively. The majority (79%) of the primed individuals achieved seroprotective MN antibody titers at 7 days after the first dose of the IIV. All LAIV-primed volunteers had MN titers ? 1:40 by Day 28 after one dose of IIV, whereas only 58% subjects from the naïve control group developed similar immune responses at this time point. The second A(H5N1) IIV dose did not increase the immune response in the LAIV-primed group, whereas 2 doses of IIV were required for naïve volunteers to develop significant immune responses. These findings were of special significance since Russian-based LAIV technology has been licensed to WHO, through whom the vaccine has been provided to vaccine manufacturers in India, China and Thailand - countries particularly vulnerable to a pandemic influenza. The results of our study will be useful to inform the development of vaccination strategies in these countries in the event of a pandemic.

Dataset Information

SIMON, an Automated Machine Learning System, Reveals Immune Signatures of Influenza Vaccine Responses.

Publications

SIMON, an Automated Machine Learning System, Reveals Immune Signatures of Influenza Vaccine Responses.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets