利用遗传与临床数据的机器学习识别可稳健预测类风湿关节炎甲氨蝶呤应答的特征标签
本研究将接受甲氨蝶呤治疗的类风湿关节炎患者按4:1分为训练集与测试集,通过递归特征消除与随机森林分类器从可能功能性单核苷酸多态性及非遗传因素中筛选特征,最终特征集含56个可能功能性单核苷酸多态性及5个非遗传因素,在测试集中曲线下面积为0.751-0.826。
要点
- 最终特征集含56个可能功能性单核苷酸多态性及5个非遗传因素。
- 训练集曲线下面积0.855-0.916,敏感度0.715-0.892,特异度0.733-0.862。
- 未见测试集曲线下面积0.751-0.826,敏感度0.581-0.839,特异度0.641-0.923。
核心结论
- 本研究通过结合生物学意义与机器学习特征筛选的新策略,识别出类风湿关节炎患者甲氨蝶呤应答的敏感且特异的预测因子,这些预测因子可能有助于改善类风湿关节炎管理中的治疗决策。
研究设计与样本
- 接受甲氨蝶呤治疗且应答已知的类风湿关节炎患者按4:1比例分为训练集与测试集。
- 从患者外显子组中,通过机器学习使用递归特征消除与交叉验证并结合随机森林分类器,从可能功能性单核苷酸多态性及非遗传因素中识别候选特征。
- 在训练队列的随机子集上重复特征选择,将共识特征组装为最终特征集,并用六种机器学习分类器评估其预测潜力。
主要发现
- 最终特征集包含56个可能功能性单核苷酸多态性及5个非遗传因素。
- 多数可能功能性单核苷酸多态性位于与类风湿关节炎发病机制或甲氨蝶呤作用相关的通路中,并预测可调节基因表达。
- 在六种机器学习分类器中训练时,训练集曲线下面积0.855-0.916,敏感度0.715-0.892,特异度0.733-0.862。
- 在未见测试集中曲线下面积0.751-0.826,敏感度0.581-0.839,特异度0.641-0.923。
关键数值与推荐
- 最终特征集含56个可能功能性单核苷酸多态性及5个非遗传因素。
- 训练集曲线下面积0.855-0.916,敏感度0.715-0.892,特异度0.733-0.862。
- 测试集曲线下面积0.751-0.826,敏感度0.581-0.839,特异度0.641-0.923。
临床怎么用
- 原文未据此给出推荐。
局限性
- 原文未报告局限性。
以上解读依据论文摘要(全文未开放获取或许可不允许翻译转载)。
原文
OBJECTIVE: To develop a hypothesis-free model that best predicts response to MTX drug in RA patients utilizing biologically meaningful genetic feature selection of potentially functional single nucleotide polymorphisms (pfSNPs) through robust machine learning (ML) feature selection methods. METHODS: MTX-treated RA patients with known response were divided in a 4:1 ratio into training and test sets. From the patients' exomes, potential features for classifier prediction were identified from pfSNPs and non-genetic factors through ML using recursive feature elimination with cross-validation incorporating the random forest classifier. Feature selection was repeated on random subsets of the training cohort, and consensus features were assembled into the final feature set. This feature set was evaluated for predictive potential using six ML classifiers, first by cross-validation within the training set, and finally by analysing its performance with the unseen test set. RESULTS: The final feature set contains 56 pfSNPs and five non-genetic factors. The majority of these pfSNPs are located in pathways related to RA pathogenesis or MTX action and are predicted to modulate gene expression. When used for training in six ML classifiers, performance was good in both the training set (area under the curve: 0.855-0.916; sensitivity: 0.715-0.892; and specificity: 0.733-0.862) and the unseen test set (area under the curve: 0.751-0.826; sensitivity: 0.581-0.839; and specificity: 0.641-0.923). CONCLUSION: Sensitive and specific predictors of MTX response in RA patients were identified in this study through a novel strategy combining biologically meaningful and machine learning feature selection and training. These predictors may facilitate better treatment decision-making in RA management.
引用
- Lim LJ, Lim AJW, Ooi BNS, et al. Machine learning using genetic and clinical data identifies a signature that robustly predicts methotrexate response in rheumatoid arthritis. Rheumatology (Oxford). 2022. PMID: 35094058. doi:10.1093/rheumatology/keac032
供执业医师工作参考,不替代临床判断;数值与分级以指南原文为准。