乙型肝炎病毒的准物种模式通过深度测序和机器学习预测肝

StephenW

Quasispecies pattern of hepatitis B virus predicts hepatocellular carcinoma using deep-sequencing and machine learning
Shipeng Chen  1 , Zihan Zhang  2 , Ying Wang  1 , Meng Fang  1 , Jun Zhou  1 , Ya Li  3 , Erhei Dai  4 , Zhaolei Feng  5 , Hao Wang  6 , Zaixing Yang  7 , Yongwei Li  8 , Xianzhang Huang  9 , Jian'an Jia  10 , Shuang Li  11 , Chenjun Huang  1 , Lin Tong  1 , Xiao Xiao  1 , Yutong He  1 , Yong Duan  3 , Shanfeng Zhu  2 , Chunfang Gao  1
Affiliations
Affiliations

    1
    Department of Laboratory Medicine, Shanghai Eastern Hepatobiliary Surgery Hospital, Shanghai, China.
    2
    Shanghai Key Lab of Intelligent Information Processing, School of Computer Science and ISTBI, Fudan University, Shanghai, China.
    3
    Department of Laboratory Medicine, The First Affiliated Hospital of Kunming Medical University, Yunnan, China.
    4
    Department of Laboratory Medicine, the Fifth Hospital of Shijiazhuang, Hebei Medical University, Hebei, China.
    5
    Department of Laboratory Medicine, Jinan infectious Disease Hospital, Shandong, China.
    6
    Department of Laboratory Medicine, Shanghai Changzheng Hospital, Shanghai, China.
    7
    Department of Laboratory Medicine, Taizhou First People's Hospital, Zhejiang, China.
    8
    Department of Laboratory Medicine, Henan Province Hospital of Traditional Chinese Medicine, Henan, China.
    9
    Department of Laboratory Medicine, The Second Affiliated Hospital of Guangzhou University of Chinese Medicine, Guangdong, China.
    10
    Department of Laboratory Medicine, Hospital of the Joint Logistics Support Force of the Chinese People's Liberation Army, Anhui, China.
    11
    Department of infectious diseases, The First Affiliated Hospital of Nanjing Medical University, Jiangsu, China.

    PMID: 33049037 DOI: 10.1093/infdis/jiaa647

Abstract

Background: Hepatitis B virus (HBV) infection is one of the main leading causes of hepatocellular carcinoma (HCC) worldwide. However, how reverse transcriptase (rt) gene contributes to HCC progression remains uncertain.

Methods: We enrolled a total of 307 chronic hepatitis B (CHB) and 237 HBV related HCC patients from 13 medical centers. Sequence features comprised multi-dimensional attributes of rt nucleic acid and rt/s amino acid sequences. Machine learning (ML) models were used to establish HCC predictive algorithms. Model performances were tested in the training and independent validation cohorts using receiver operating characteristic (ROC) and calibration plots.

Results: Random forest (RF) model based on combined metrics (10 features) demonstrated the best predictive performances in both cross and independent validation (RFAUC=0.96, RFACC=0.90), irrespective of HBV genotypes and sequencing depth. Moreover, HCC risk score for individuals obtained from the RF model (AUC =0.966, 95% CI=0.922-0.989) outperformed α-fetal protein (AUC=0.713, 95% CI=0.632-0.784) in identifying HCC from CHB patients.

Conclusions: Our study provides evidence for the first time that HBV rt sequences contain vital HBV quasispecies features in predicting HCC. Integrating deep sequencing with feature extraction and ML models benefits the longitudinal surveillance of CHB and HCC risk assessment.

Keywords: algorithm; hepatitis B virus (HBV); hepatocellular carcinoma (HCC); machine learning (ML); next-generation sequencing (NGS).

© The Author(s) 2020. Published by Oxford University Press for the Infectious Diseases Society of America. All rights reserved. For permissions, e-mail: [email protected].

StephenW

乙型肝炎病毒的准物种模式通过深度测序和机器学习预测肝细胞癌
陈仕鹏1，张子涵2，王颖1，孟芳1，君州1，亚莉3，戴黑河4，赵雷锋5，王皓6，杨子星7，李永伟8，黄宪章9，建'安家10，双力11，黄晨军1，林彤1，肖小1，河童1，雍段3，朱善峰2，高春芳1
隶属关系
隶属关系

    1个
    上海市东方肝胆外科医院检验医学科，上海。
    2
    复旦大学计算机科学与技术学院，上海智能信息处理重点实验室，上海
    3
    昆明医科大学附属第一医院检验医学科，云南。
    4
    河北医科大学附属石家庄市第五医院检验医学科
    5
    山东省济南市传染病医院检验科。
    6
    上海长征医院检验科，上海
    7
    浙江省台州市第一人民医院检验科。
    8
    河南省中医院检验科，河南。
    9
    广州中医药大学附属第二医院检验医学科
    10
    中国人民解放军联合后勤保障部队医院检验科，安徽。
    11
    南京医科大学附属第一医院感染科

    PMID：33049037 DOI：10.1093 / infdis / jiaa647

抽象

背景：乙型肝炎病毒（HBV）感染是全球肝细胞癌（HCC）的主要主要原因之一。然而，逆转录酶（rt）基因如何促进肝癌的进展尚不确定。

方法：我们招募了来自13个医疗中心的307例慢性乙型肝炎（CHB）和237例HBV相关的HCC患者。序列特征包括rt核酸和rt / s氨基酸序列的多维属性。机器学习（ML）模型用于建立HCC预测算法。使用接收器操作特征（ROC）和校准图在训练和独立验证队列中测试了模型性能。

结果：基于合并指标（10个特征）的随机森林（RF）模型在交叉验证和独立验证中均表现出最佳的预测性能（RFAUC = 0.96，RFACC = 0.90），而与HBV基因型和测序深度无关。此外，从RF模型（AUC = 0.966，95％CI = 0.922-0.989）获得的个体的HCC风险评分在从CHB患者中识别HCC方面优于α-胎儿蛋白（AUC = 0.713，95％CI = 0.632-0.784）。

结论：我们的研究首次提供了证据，证明HBV rt序列在预测HCC中包含重要的HBV准种特征。将深度测序与特征提取和ML模型集成在一起，可对CHB和HCC风险评估进行纵向监控。

关键词：算法乙型肝炎病毒（HBV）；肝细胞癌（HCC）；机器学习（ML）;下一代测序（NGS）。

©2020作者。牛津大学出版社，美国传染病学会出版。版权所有。有关权限，请发送电子邮件至：[email protected]。