清晨好,您是今天最早来到科研通的研友!由于当前在线用户较少,发布求助请尽量完整地填写文献信息,科研通机器人24小时在线,伴您科研之路漫漫前行!

Evaluating Machine Learning Models for Molecular Property Prediction: Performance and Robustness on Out-of-Distribution Data

机器学习 计算机科学 聚类分析 人工智能 稳健性(进化) 适用范围 人工神经网络 训练集 图形 财产(哲学) 相似性(几何) 数据挖掘 相关性 数据点 数据建模 领域(数学分析) 任务(项目管理) 聚类系数 线性模型 领域知识 标记数据 网络拓扑 监督学习 对比度(视觉) 试验数据 概率分布 随机森林 实验数据 图论 学习迁移 深度学习
作者
Hosein Fooladi,Thi Ngoc Lan Vu,Miriam Mathea,Johannes Kirchmair
出处
期刊:Journal of Chemical Information and Modeling [American Chemical Society]
卷期号:65 (19): 9871-9891 被引量:72
标识
DOI:10.1021/acs.jcim.5c00475
摘要

High Resolution Image Download MS PowerPoint Slide Today, machine learning models are employed extensively to predict the physicochemical and biological properties of molecules. Their performance is typically evaluated on in-distribution (ID) data, i.e., data originating from the same distribution as the training data. However, the real-world applications of such models often involve molecules that are more distant from the training data, necessitating the assessment of their performance on out-of-distribution (OOD) data. In this work, we investigate and evaluate the performance of 14 machine learning models, including classical approaches like random forests, as well as graph neural network (GNN) methods, such as message-passing graph neural networks, across eight data sets using ten splitting strategies for OOD data generation. First, we investigate what constitutes OOD data in the molecular domain for bioactivity and ADMET prediction tasks. In contrast to the common point of view, we show that both classical machine learning and GNN models work well (not substantially different from random splitting) on data split based on Bemis-Murcko scaffolds. Splitting based on chemical similarity clustering (UMAP-based clustering using ECFP4 fingerprints) poses the most challenging task for both types of models. Second, we investigate the extent to which ID and OOD performance have a positive linear relationship. If a positive correlation holds, models with the best performance on the ID data can be selected with the promise of having the best performance on OOD data. We show that the strength of this linear relationship is strongly related to how the OOD data is generated, i.e., which splitting strategies are used for generating OOD data. While the correlation between ID and OOD performance for scaffold splitting is strong (Pearson’s r ∼ 0.9), this correlation decreases significantly for all the cluster-based splitting (Pearson’s r ∼ 0.4). Therefore, the relationship can be more nuanced, and a strong positive correlation is not guaranteed for all OOD scenarios. These findings suggest that OOD performance evaluation and model selection should be carefully aligned with the intended application domain.
最长约 10秒,即可获得该文献文件

科研通智能强力驱动
Strongly Powered by AbleSci AI
科研通是完全免费的文献互助平台,具备全网最快的应助速度,最高的求助完成率。 对每一个文献求助,科研通都将尽心尽力,给求助人一个满意的交代。
实时播报
2秒前
喷火球发布了新的文献求助100
7秒前
开心的寄柔完成签到,获得积分10
9秒前
blenx完成签到,获得积分10
22秒前
33秒前
迷路万天完成签到,获得积分10
38秒前
39秒前
41秒前
隐形忆丹完成签到,获得积分10
49秒前
52秒前
55秒前
Jun的应助被JINX采纳,获得10
55秒前
可靠的夕阳完成签到,获得积分10
1分钟前
ww发布了新的文献求助10
1分钟前
ww完成签到,获得积分10
1分钟前
Criminology34的应助被科研通管家采纳,获得10
1分钟前
Criminology34的应助被科研通管家采纳,获得10
1分钟前
null的应助被科研通管家采纳,获得10
1分钟前
Criminology34的应助被科研通管家采纳,获得10
1分钟前
精明魔镜完成签到,获得积分10
1分钟前
五月完成签到,获得积分10
1分钟前
1分钟前
1分钟前
善良凝荷完成签到,获得积分10
1分钟前
2分钟前
科研通AI6.3的应助被喷火球采纳,获得10
2分钟前
adgn发布了新的文献求助10
2分钟前
2分钟前
标致茈完成签到,获得积分10
2分钟前
2分钟前
小马甲的应助被TingtingGZ采纳,获得10
2分钟前
sfwrbh完成签到,获得积分10
2分钟前
2分钟前
激昂的慕山完成签到,获得积分10
2分钟前
TingtingGZ发布了新的文献求助10
2分钟前
2分钟前
Hello的应助被宜兴彭于晏采纳,获得10
2分钟前
平常以云完成签到 ,获得积分10
2分钟前
CC完成签到,获得积分10
2分钟前
2分钟前
高分求助中
(应助此贴封号)【重要!!请各用户(尤其是新用户)详细阅读】【科研通的精品贴汇总】 10000
拉伸高分子熔体中缠结行为的模拟和小角中子散射研究 1700
The World Atlas of Wine 9th edition 1000
奈特基础组织学:配套组织病理学(第4版)(Netter's Essential Histology With Correlated Histopathology) 800
John McDowells Bildungstheorie 800
八旗制度の研究 600
plantTFDB 508
热门求助领域 (近24小时)
化学 材料科学 医学 生物 纳米技术 计算机科学 工程类 有机化学 化学工程 生物化学 物理 复合材料 内科学 催化作用 光电子学 人工智能 细胞生物学 无机化学 物理化学 数学
热门帖子
关注 科研通微信公众号,转发送积分 8173161
求助须知:如何正确求助?哪些是违规求助? 9565558
关于积分的说明 21780458
捐赠科研通 7767231
什么是DOI,文献DOI怎么找? 3431542
关于科研通互助平台的介绍 2555697
邀请新用户注册赠送积分活动 2476682