arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1809.09143cs.LGq-bio.QMstat.ML
收藏

EpiRL:一个促进上位性检测的强化学习代理

EpiRL: A Reinforcement Learning Agent to Facilitate Epistasis Detection

Courant Institute(库朗研究所);New York University(纽约大学);Tandon School of Engineering(坦登工程学院)

Kexin Huang, Rodrigo Nogueira

摘要:

上位性(基因-基因相互作用)对预测遗传疾病至关重要。我们的工作通过将其建模为一步马尔可夫决策过程,其中状态是基因组数据,动作是相互作用的基因,奖励是所选动作的相互作用度量,来解决以前的上位性检测工作所面临的计算挑战。一个强化学习代理使用策略梯度方法,然后学习发现一组高度相互作用的基因。

英文摘要:

Epistasis (gene-gene interaction) is crucial to predicting genetic disease. Our work tackles the computational challenges faced by previous works in epistasis detection by modeling it as a one-step Markov Decision Process where the state is genome data, the actions are the interacted genes, and the reward is an interaction measurement for the selected actions. A reinforcement learning agent using policy gradient method then learns to discover a set of highly interacted genes.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑