Molecule Identification with Rotational Spectroscopy and Probabilistic Deep Learning
Molecule Identification with Rotational Spectroscopy and Probabilistic Deep Learning
复制标题
DOI：
10.1021/acs.jpca.0c01376
复制
发表时间：
2020-04-16
影响因子：
2.9
通讯作者：
Lee, Kin Long Kelvin
中科院分区：
文献类型：
--
作者：
McCarthy, Michael;Lee, Kin Long Kelvin
提出了一个概念验证框架，用于使用实验旋转数据和概率深度学习来识别未知元素组成和结构的分子。使用实验确定的最小输入数据集，我们描述了四个神经网络架构，产生信息，以帮助识别未知分子。第一种架构将光谱参数转换为库仑矩阵本征谱，作为恢复编码在旋转光谱中的化学和结构信息的方法。随后，三个深度学习网络使用本征谱来限制化学计量的范围，生成SMILES字符串，并预测分子中最可能存在的官能团。在每个模型中，我们利用dropout层作为贝叶斯采样的近似，然后从其他确定性模型中生成概率预测。这些模型在中等大小的理论数据集上进行训练，该数据集包括类似于在omega B 97 X-D/6-31+G（d）理论水平上优化的83 000个独特的有机分子（18至180 amu之间），其中光谱常数的理论不确定性得到了很好的理解并用于进一步增强训练。由于化学和结构性质强烈依赖于分子组成，我们将数据集分为四组，分别对应于纯烃、含氧物质、含氮物质以及含氧和含氮物质，用其中一个类别训练每种类型的网络，从而在每个分子领域内创建“专家”。我们演示了如何将这些模型用于四个分子的实际推理，并讨论了我们的方法的优点和缺点以及这些架构可以采取的未来方向。
A proof-of-concept framework for identifying molecules of unknown elemental composition and structure using experimental rotational data and probabilistic deep learning is presented. Using a minimal set of input data determined experimentally, we describe four neural network architectures that yield information to assist in the identification of an unknown molecule. The first architecture translates spectroscopic parameters into Coulomb matrix eigenspectra as a method of recovering chemical and structural information encoded in the rotational spectrum. The eigenspectrum is subsequently used by three deep learning networks to constrain the range of stoichiometries, generate SMILES strings, and predict the most likely functional groups present in the molecule. In each model, we utilize dropout layers as an approximation to Bayesian sampling, which subsequently generates probabilistic predictions from otherwise deterministic models. These models are trained on a modestly sized theoretical dataset comprising , similar to 83 000 unique organic molecules (between 18 and 180 amu) optimized at the omega B97X-D/6-31+G(d) level of theory, where the theoretical uncertainties of the spectoscopic constants are well-understood and used to further augment training. Since chemical and structural properties depend strongly on molecular composition, we divided the dataset into four groups corresponding to pure hydrocarbons, oxygen-bearing species, nitrogen-bearing species, and both oxygen- and nitrogen-bearing species, training each type of network with one of these categories, thus creating "experts" within each domain of molecules. We demonstrate how these models can then be used for practical inference on four molecules and discuss both the strengths and shortcomings of our approach and the future directions these architectures can take.