厦门大学平台管理系统 Fan Lin--Home-- SGCLMD: Signed graph-based contrastive learning model for predicting somatic mutation-drug association

Fan Lin Associate Professor

林凡，理学博士，现担任厦门大学平潭研究院执行院长、硕/博士生导师、软件工程系副教授、IEEE高级会员；指导的已毕业硕士研究生22人、博士5人。主要研究方向包括：强化学习、推荐算法、生成模型近年来致力于研究生成式模型方法的深层次应用。包括基于生成模型的时间序列预测分析、数字虚拟人、RLHF应用教育、金融交叉领域的研究。主持教育交叉学科国家自然科学基金面上项目、主持国家科技支撑计划、主持科技部中小企业创新基金...Detials

Paper Publications Current position: Home > Scientific Research > Paper Publications

SGCLMD: Signed graph-based contrastive learning model for predicting somatic mutation-drug association

Release time:2026-10-16 Hits:

Impact Factor:6.305
DOI number:10.1016/j.compbiomed.2025.110067
Affiliation of Author(s):厦门大学信息学院
Journal:Computers in Biology and Medicine
Key Words:Graph neural network; Multi-view contrastive learning; Signed graph; Somatic mutation-drug association
Abstract:中文概述：提出SGCLMD突变—药物关联预测模型，通过随机扰动构造有符号图的增强视图，并以多视图对比学习获得节点表示，用于区分突变与药物之间的敏感或耐药关联。
Note:文章编号：110067；日期口径：正式在线发表日；卷期2025-05；IF及分区沿用简历，统计年份未提供；通讯作者沿用简历；来源：https://doi.org/10.1016/j.compbiomed.2025.110067 https://pubmed.ncbi.nlm.nih.gov/40147185/；核实日期2026-10-16；摘要为据原摘要撰写的中文概述
First Author:Xiaosong Wang
Correspondence Author:Fan Lin
Co-author:Haisong Feng, Yilei Zhang,Fan Lin
Volume:190
ISSN No.:0010-4825
Translation or Not:no
Date of Publication:2025-03-26

Pre One: Adaptive Bidirectional State Space Model for High-frequency Portfolio Management

Next One: Knowledge-aware reasoning with self-supervised reinforcement learning for explainable recommendation in MOOCs