许鹏^{1, 3},
王兵²,
方刚¹,
石晓龙¹,
刘文斌^{1, 2,,}
1.广州大学计算科技研究院 广州 510006
2.温州大学计算机与人工智能学院 温州 325035
3.黔南民族师范学院计算机与信息学院 都匀 558000
基金项目:国家重点研发计划(2019YFA0706402)，国家自然科学基金(61572367, 61573017, 61972107, 61972109)

详细信息

作者简介:许鹏：男，1986年生，博士后，研究方向为生物信息学
王兵：男，1993年生，硕士生，研究方向为生物信息学
方刚：男，1969年生，教授，研究方向为生物信息学
石晓龙：男，1975年生，教授，研究方向为生物信息学
刘文斌：男，1969年生，教授，研究方向为生物信息学

通讯作者:刘文斌　wbliu6910@126.com

中图分类号:TP391

计量

文章访问数:1055
HTML全文浏览量:309
PDF下载量:46
被引次数:0

出版历程

收稿日期:2019-11-01
修回日期:2020-05-10
网络出版日期:2020-05-23
刊出日期:2020-06-22

Analysis of Breast Cancer Subtypes Prediction Based on Alternative Splicing Disorders

Peng XU^{1, 3},
Bing WANG²,
Gang FANG¹,
Xiaolong SHI¹,
Wenbin LIU^{1, 2,,}
1. Institute of Computing Science and Technology, Guangzhou University, Guangzhou 510006, China
2. College of Computer Science and Artificial Intelligence, Wenzhou University, Wenzhou 325035, China
3. School of Computer Science and Information Technology, Qiannan Normal University for Nationalities, Duyun 558000, China
Funds:The National Key R&D Program of China (2019YFA0706402), The National Natural Science Foundation of China (61572367, 61573017, 61972107, 61972109)


摘要
摘要:可变剪接与多种复杂疾病的发生、发展存在密切的联系，包括肿瘤在内的多种疾病的产生往往伴随着可变剪接的紊乱发生。现有的乳腺癌亚型分析主要是基于单个剪接异构体出发，缺少考虑亚型之间由于可变剪接紊乱造成剪接异构体在整体分布上的差异。因此该文提出了基于可变剪接紊乱的乳腺癌亚型预测方法，主要使用Jensen-Shannon(JS)散度来找寻亚型之间的可变剪接紊乱差异较大的基因，并构建反向传播(BP)神经网络模型对乳腺癌亚型进行分类。结果表明，该方法不仅能有效发现肿瘤异质性分子，在乳腺癌亚型分类方面也有较好的识别结果，其平均F1值达到0.89，且能为患者提供个性化乳腺癌亚型药物推荐。该文的研究将有效促进基于可变剪接紊乱的乳腺癌亚型研究的发展。
关键词:乳腺癌亚型预测/
可变剪接/
Jensen-Shannon散度/
反向传播神经网络/
药物推荐
Abstract:Alternative splicing is closely related to the occurrence and development of a variety of complex diseases, the emergence of various diseases including tumors is often accompanied by the occurrence of alternative splicing disorders. The existing analysis of breast cancer subtypes is mainly based on single splicing isoform, and the difference in the overall distribution of splicing isoforms caused by alternative splicing disorders among subtypes is not considered. Therefore, a prediction method of breast cancer subtypes based on alternative splicing disorders is proposed, which mainly uses Jensen-Shannon(JS) divergence to find genes with large differences in alternative splicing disorders between subtypes, then constructes Back Propagation(BP) neural network model to classify breast cancer subtypes. The results show that this method could not only effectively detect tumor heterogeneous molecules, but also had good identification results in the classification of breast cancer subtypes, with an average F1-score of 0.89, and could provide personalized drug recommendations for patients with breast cancer subtypes. This study will effectively promote the development of breast cancer subtypes based on alternative splicing disorders.
Key words:Breast cancer subtypes predicion/
Alternative splicing/
Jensen-Shannon(JS) divergence/
Back Propagation(BP) neural network/
Drug recommendation



PDF全文下载地址:

https://jeit.ac.cn/article/exportPdf?id=ae0ae7c5-69af-4479-b0cf-530855efc16b