A step toward a reinforcement learning de novo genome assembler
Authors:
Kleber Padovani,
Roberto Xavier,
Rafael Cabral Borges,
Andre Carvalho,
Anna Reali,
Annie Chateau,
Ronnie Alves
Abstract:
De novo genome assembly is a relevant but computationally complex task in genomics. Although de novo assemblers have been used successfully in several genomics projects, there is still no 'best assembler', and the choice and setup of assemblers still rely on bioinformatics experts. Thus, as with other computationally complex problems, machine learning may emerge as an alternative (or complementary…
▽ More
De novo genome assembly is a relevant but computationally complex task in genomics. Although de novo assemblers have been used successfully in several genomics projects, there is still no 'best assembler', and the choice and setup of assemblers still rely on bioinformatics experts. Thus, as with other computationally complex problems, machine learning may emerge as an alternative (or complementary) way for developing more accurate and automated assemblers. Reinforcement learning has proven promising for solving complex activities without supervision - such games - and there is a pressing need to understand the limits of this approach to 'real' problems, such as the DFA problem. This study aimed to shed light on the application of machine learning, using reinforcement learning (RL), in genome assembly. We expanded upon the sole previous approach found in the literature to solve this problem by carefully exploring the learning aspects of the proposed intelligent agent, which uses the Q-learning algorithm, and we provided insights for the next steps of automated genome assembly development. We improved the reward system and optimized the exploration of the state space based on pruning and in collaboration with evolutionary computing. We tested the new approaches on 23 new larger environments, which are all available on the internet. Our results suggest consistent performance progress; however, we also found limitations, especially concerning the high dimensionality of state and action spaces. Finally, we discuss paths for achieving efficient and automated genome assembly in real scenarios considering successful RL applications - including deep reinforcement learning.
△ Less
Submitted 7 March, 2024; v1 submitted 2 February, 2021;
originally announced February 2021.
Prime Focus Spectrograph (PFS) for the Subaru Telescope: Overview, recent progress, and future perspectives
Authors:
Naoyuki Tamura,
Naruhisa Takato,
Atsushi Shimono,
Yuki Moritani,
Kiyoto Yabe,
Yuki Ishizuka,
Akitoshi Ueda,
Yukiko Kamata,
Hrand Aghazarian,
Stephane Arnouts,
Gabriel Barban,
Robert H. Barkhouser,
Renato C. Borges,
David F. Braun,
Michael A. Carr,
Pierre-Yves Chabaud,
Yin-Chang Chang,
Hsin-Yo Chen,
Masashi Chiba,
Richard C. Y. Chou,
You-Hua Chu,
Judith G. Cohen,
Rodrigo P. de Almeida,
Antonio C. de Oliveira,
Ligia S. de Oliveira
, et al. (75 additional authors not shown)
Abstract:
PFS (Prime Focus Spectrograph), a next generation facility instrument on the 8.2-meter Subaru Telescope, is a very wide-field, massively multiplexed, optical and near-infrared spectrograph. Exploiting the Subaru prime focus, 2394 reconfigurable fibers will be distributed over the 1.3 deg field of view. The spectrograph has been designed with 3 arms of blue, red, and near-infrared cameras to simult…
▽ More
PFS (Prime Focus Spectrograph), a next generation facility instrument on the 8.2-meter Subaru Telescope, is a very wide-field, massively multiplexed, optical and near-infrared spectrograph. Exploiting the Subaru prime focus, 2394 reconfigurable fibers will be distributed over the 1.3 deg field of view. The spectrograph has been designed with 3 arms of blue, red, and near-infrared cameras to simultaneously observe spectra from 380nm to 1260nm in one exposure at a resolution of ~1.6-2.7A. An international collaboration is developing this instrument under the initiative of Kavli IPMU. The project is now going into the construction phase aiming at undertaking system integration in 2017-2018 and subsequently carrying out engineering operations in 2018-2019. This article gives an overview of the instrument, current project status and future paths forward.
△ Less
Submitted 3 August, 2016;
originally announced August 2016.
