Leffingwell Odor Dataset

NOTE: It's easier to download this dataset from pyrfume. Here's how: # First install pyrfume in your Python environment. This can be done easily with pip.
# pip install pyrfume

import pyrfume
molecules = pyrfume.load_data('leffingwell/molecules.csv', remote=True)
behavior = pyrfume.load_data('leffingwell/behavior.csv', remote=True)
# e.g. to count the number of molecules with each descriptor
behavior.sum().sort_values(ascending=False).astype(int)

Predicting properties of molecules is an area of growing research in machine learning, particularly as models for learning from graph-valued inputs improve in sophistication and robustness. A molecular property prediction problem that has received comparatively little attention during this surge in research activity is building Structure-Odor Relationships (SOR) models (as opposed to Quantitative Structure-Activity Relationships, a term from medicinal chemistry). This is a 70+ year-old problem straddling chemistry, physics, neuroscience, and machine learning. To spur development on the SOR problem, we curated and cleaned a dataset of 3523 molecules associated with expert-labeled odor descriptors from the Leffingwell PMP 2001 database. We provide featurizations of all molecules in the dataset using bit-based and count-based fingerprints, Mordred molecular descriptors, and the embeddings from our trained GNN model (Sanchez-Lengeling et al., 2019). This dataset is comprised of two files: leffingwell_data.csv: this contains molecular structures, and what they smell like, along with train, test, and cross-validation splits. More detail on the file structure is found in leffingwell_readme.pdf. leffingwell_embeddings.npz: this contains several featurizations of the molecules in the dataset. leffingwell_readme.pdf: a more detailed description of the data and its provenance, including expected performance metrics. LICENSE: a copy of the CC-BY-NC license language. The dataset, and all associated features, is freely available for research use under the CC-BY-NC license. If you use the data in a publication, please cite: @article{sanchez2019machine,
title={Machine learning for scent: Learning generalizable perceptual representations of small molecules},
author={Sanchez-Lengeling, Benjamin and Wei, Jennifer N and Lee, Brian K and Gerkin, Richard C and Aspuru-Guzik, Al{\'a}n and Wiltschko, Alexander B},
journal={arXiv preprint arXiv:1910.10685},
year={2019}
}

注意：您可通过pyrfume平台更便捷地下载本数据集，操作步骤如下：
# 首先在Python环境中安装pyrfume，可通过pip快速完成安装：
pip install pyrfume

import pyrfume
molecules = pyrfume.load_data('leffingwell/molecules.csv', remote=True)
behavior = pyrfume.load_data('leffingwell/behavior.csv', remote=True)
# 示例：统计每个气味描述符对应的分子数量
behavior.sum().sort_values(ascending=False).astype(int)

预测分子属性是机器学习领域日益受到关注的研究方向，尤其是当基于图结构输入的学习模型在复杂度与鲁棒性上持续提升之时。在这场研究热潮中，相对未获得足够关注的一类分子属性预测问题，是构建结构-气味关系（Structure-Odor Relationships, SOR）模型（区别于药物化学领域的定量构效关系（Quantitative Structure-Activity Relationships））。该课题横跨化学、物理学、神经科学与机器学习等多个领域，已有70余年的研究历史。

为推动结构-气味关系问题的研究进展，我们从Leffingwell PMP 2001数据库中整理并清洗了包含3523个分子的数据集，所有分子均附带专家标注的气味描述符。我们为数据集中的全部分子提供了多种特征化方式：基于位的指纹（bit-based fingerprints）、基于计数的指纹（count-based fingerprints）、Mordred分子描述符，以及我们训练的图神经网络（Graph Neural Network, GNN）模型生成的嵌入向量（Sanchez-Lengeling等，2019）。

本数据集由以下文件组成：
1. `leffingwell_data.csv`：包含分子结构、气味描述信息，以及训练集、测试集与交叉验证的划分方案。更多文件结构细节可参阅`leffingwell_readme.pdf`。
2. `leffingwell_embeddings.npz`：包含数据集中所有分子的多种特征化结果。
3. `leffingwell_readme.pdf`：对数据集及其来源的详细说明，包含预期性能指标。
4. `LICENSE`：CC-BY-NC许可协议的文本副本。

本数据集及所有关联特征均遵循CC-BY-NC许可协议，可免费用于科研用途。若您在发表的学术成果中使用本数据集，请引用如下文献：
bibtex
@article{sanchez2019machine,
title={Machine learning for scent: Learning generalizable perceptual representations of small molecules},
author={Sanchez-Lengeling, Benjamin and Wei, Jennifer N and Lee, Brian K and Gerkin, Richard C and Aspuru-Guzik, Alán and Wiltschko, Alexander B},
journal={arXiv preprint arXiv:1910.10685},
year={2019}
}

搜集汇总

背景与挑战

背景概述

Leffingwell Odor Dataset是一个用于机器学习研究的气味分子数据集，包含3523个分子及其专家标记的气味描述符，源自Leffingwell PMP 2001数据库。数据集提供了多种分子特征化方法，如指纹、Mordred描述符和GNN嵌入，旨在促进结构-气味关系（SOR）模型的开发。该数据集遵循CC-BY-NC许可，仅供非商业研究使用。