MutaGAN: A Seq2seq GAN Framework to Predict Mutations of Evolving Protein Populations
Authors:
Daniel S. Berman,
Craig Howser,
Thomas Mehoke,
Jared D. Evans
Abstract:
The ability to predict the evolution of a pathogen would significantly improve the ability to control, prevent, and treat disease. Despite significant progress in other problem spaces, deep learning has yet to contribute to the issue of predicting mutations of evolving populations. To address this gap, we developed a novel machine learning framework using generative adversarial networks (GANs) wit…
▽ More
The ability to predict the evolution of a pathogen would significantly improve the ability to control, prevent, and treat disease. Despite significant progress in other problem spaces, deep learning has yet to contribute to the issue of predicting mutations of evolving populations. To address this gap, we developed a novel machine learning framework using generative adversarial networks (GANs) with recurrent neural networks (RNNs) to accurately predict genetic mutations and evolution of future biological populations. Using a generalized time-reversible phylogenetic model of protein evolution with bootstrapped maximum likelihood tree estimation, we trained a sequence-to-sequence generator within an adversarial framework, named MutaGAN, to generate complete protein sequences augmented with possible mutations of future virus populations. Influenza virus sequences were identified as an ideal test case for this deep learning framework because it is a significant human pathogen with new strains emerging annually and global surveillance efforts have generated a large amount of publicly available data from the National Center for Biotechnology Information's (NCBI) Influenza Virus Resource (IVR). MutaGAN generated "child" sequences from a given "parent" protein sequence with a median Levenshtein distance of 2.00 amino acids. Additionally, the generator was able to augment the majority of parent proteins with at least one mutation identified within the global influenza virus population. These results demonstrate the power of the MutaGAN framework to aid in pathogen forecasting with implications for broad utility in evolutionary prediction for any protein population.
△ Less
Submitted 26 August, 2020;
originally announced August 2020.
Managing Distributed Software Development in the Virtual Astronomical Observatory
Authors:
Janet D. Evans,
Raymond L. Plante,
Nina Bonaventura,
Ivo Busko,
Mark Cresitello-Dittmar,
Raffaele D'Abrusco,
Stephen Doe,
Rick Ebert,
Omar Laurino,
Olga Pevunova,
Brian Refsdal,
Brian Thomas
Abstract:
The U.S. Virtual Astronomical Observatory (VAO) is a product-driven organization that provides new scientific research capabilities to the astronomical community. Software development for the VAO follows a lightweight framework that guides development of science applications and infrastructure. Challenges to be overcome include distributed development teams, part-time efforts, and highly constrain…
▽ More
The U.S. Virtual Astronomical Observatory (VAO) is a product-driven organization that provides new scientific research capabilities to the astronomical community. Software development for the VAO follows a lightweight framework that guides development of science applications and infrastructure. Challenges to be overcome include distributed development teams, part-time efforts, and highly constrained schedules. We describe the process we followed to conquer these challenges while developing Iris, the VAO application for analysis of 1-D astronomical spectral energy distributions (SEDs). Iris was successfully built and released in less than a year with a team distributed across four institutions. The project followed existing International Virtual Observatory Alliance inter-operability standards for spectral data and contributed a SED library as a by-product of the project. We emphasize lessons learned that will be folded into future development efforts. In our experience, a well-defined process that provides guidelines to ensure the project is cohesive and stays on track is key to success. Internal product deliveries with a planned test and feedback loop are critical. Release candidates are measured against use cases established early in the process, and provide the opportunity to assess priorities and make course corrections during development. Also key is the participation of a stakeholder such as a lead scientist who manages the technical questions, advises on priorities, and is actively involved as a lead tester. Finally, frequent scheduled communications (for example a bi-weekly tele-conference) assure issues are resolved quickly and the team is working toward a common vision
△ Less
Submitted 26 June, 2012;
originally announced June 2012.
