Sequence-to-sequence Learning For Molecular Structure Derivation From Infrared Spectra

French, Ethan

Permalink

https://hdl.handle.net/2142/116578 Copy

Description

Title

Sequence-to-sequence Learning For Molecular Structure Derivation From Infrared Spectra

Author(s)

French, Ethan

Contributor(s)

Lin, Zhou

Issue Date

2022-06-22

Keyword(s)

Mini-symposium: Machine Learning

Abstract

\begin{wrapfigure}{r}{0pt} \includegraphics[scale=0.4]{model.eps} \end{wrapfigure} Fully identifying unknown molecules via infrared spectroscopy can be a challenging task for even the most experienced researchers. Current data-driven computational methods usually identify unknown spectra by matching them against databases of known spectra. However, this method can be problematic for novel complex molecules given the relative lack of information. Deep learning provides a potential solution to this problem. Sequence-to-sequence learning has had great success in a wide range of areas such as language translation and speech recognition.\footnote{Ilya Sutskever, Oriol Vinyals, Quoc V. Le, ``Sequence to Sequence Learning with Neural Networks'', {\it NeurIPS}, {2014}, {\bf 27}.} In this work, an unsupervised sequence-to-sequence model was extended to chemical systems and used to derive complete molecular structures from infrared spectra. The model was trained on the infrared spectra of small organic molecules containing C, H, O, N, and F atoms. These molecules were represented using SELFIES, an improved version of the SMILES string molecular fingerprint descriptor.\footnote{Mario Krenn, Florian H\"{a}se, AkshatKumar Nigam, Pascal Friederich, and Alan Aspuru-Guzik, ``Self-referencing embedded strings (SELFIES): A 100\% robust molecular string representation'', {\it Mach. Learn.: Sci. Technol.} 2020, {\bf 1}, 045024} Our model is able to achieve state-of-the-art results in successfully identifying a wide variety of molecules from their infrared spectra.

Publisher

International Symposium on Molecular Spectroscopy

Type of Resource

text

Language

eng

Handle URL

https://hdl.handle.net/2142/116578&&

DOI

https://doi.org/10.15278/isms.2022.WK05

Copyright and License Information

Copyright 2022 held by the authors

Owning Collections