PALO Bounds for Reinforcement Learning In Partially Observable Stochastic Games
Document Type
Article
Publication Date
1-8-2021
School
Computing Sciences and Computer Engineering
Abstract
Following the 2010 Deepwater Horizon oil spill, extensive research has been conducted on the toxicity of oil and polycyclic aromatic hydrocarbons (PAHs) in the aquatic environment. Many studies have identified the toxicological effects of PAHs in estuarine and marine fishes, however, only recently has work begun to identify the combinatorial effect of PAHs and abiotic environmental factors such as hypoxia, salinity, and temperature. This study aims to characterize the combined effects of abiotic stressors and PAH exposure on the cardiac transcriptomes of developing Fundulus grandis larvae. In this study, F. grandis larvae were exposed to varying environmental conditions (dissolved oxygen (DO) 2, 6 ppm; temperature 20, 30 °C; and salinity 3, 30 ppt) as well as to a single concentration of high energy water accommodated fraction (HEWAF) (∑PAHs 15 ppb). Whole larvae were sampled for RNA and transcriptional changes were quantified using RNA-Seq followed by qPCR for a set of target genes. Analysis revealed that exposure to oil and abiotic stressors impacts signaling pathways associated with cardiovascular function. Specifically, combined exposures appear to reduce development of the systemic vasculature as well as strongly impact the cardiac musculature through cardiomyocyte proliferation resulting in inhibited cardiac function and modulated blood pressure maintenance. Results of this study provide a holistic view of impacts of PAHs and common environmental stressors on the cardiac system in early life stage estuarine species. To our knowledge, this study is one of the first to simultaneously manipulate oil exposure with abiotic factors (DO, salinity, temperature) and the first to analyze cardiac transcriptional responses under these co-exposures.
Publication Title
Neurocomputing
Volume
420
First Page
36
Last Page
56
Recommended Citation
Ceren, R.,
He, K.,
Doshi, P.,
Banerjee, B.
(2021). PALO Bounds for Reinforcement Learning In Partially Observable Stochastic Games. Neurocomputing, 420, 36-56.
Available at: https://aquila.usm.edu/fac_pubs/18355
COinS
Comments
© This manuscript version is made available under the CC-BY-NC-ND 4.0 license https://creativecommons.org/licenses/by-nc-nd/4.0/.