Machine learning (ML) is applied broadly in chemical properties of single values such as ionization potential and electronic affinity. While the prediction of absorption spectrum by ML method is relatively few. In our study, accurate infrared (IR) and THz spectra of common organic molecules is determined by investigating the structure-property relationship. We choose multi-layer perceptron (MLP), recurrent neural network (RNN), and convolutional neural network (CNN) as the statistical classification algorithm, in accordance with molecular descriptors topological indices, MACCS fingerprint, adjacency matrix, or Coulomb matrix. All the 360 organic molecule IR spectra are from Pacific Northwestern National Lab (PNNL) database as the training set and are assessed. THz comparisons are made with recently calculated DFT specta of fentanyl and analogues. Current work will lead to intelligent detection schemes for future opiods and derivatives.

Machine learning methods for the prediction of spectra and identification of opioids and their derivatives

Macromolecular Chemistry: The Second Century

[Author Benefits](https://scimeetings.acs.org/?utm_source=pubs_content_marketing&utm_medium=website&utm_campaign=0320_MCF_NPI_Launch_Spring_Homepage&ref=pubs_content_marketing)

[How to use SciMeetings](https://storage.googleapis.com/dl.morressier.com/acs/SciMeetings-ACSSpring2020-HowToGuide_M3020-Final.pdf)

[Frequently Asked Questions](https://storage.googleapis.com/dl.morressier.com/acs/SciMtng_ACSNMSp2020_FAQ_M3120_Final.pdf)

ACS Spring 2020 National Meeting & Expo

The official archive and repository for all ACS Meetings & Expositions abstracts and select posters or presentations.