SELFIES and the future of molecular string representations

Artificial intelligence (AI) and machine learning (ML) are expanding in popularity for broad applications to challenging tasks in chemistry and materials science. Examples include the prediction of properties, the discovery of new reaction pathways, or the design of new molecules. The machine needs to read and write fluently in a chemical language for each of these tasks. Strings are a common tool to represent molecular graphs, and the most popular molecular string representation, Smiles, has powered cheminformatics since the late 1980s. However, in the context of AI and ML in chemistry, Smiles has several shortcomings-most pertinently, most combinations of symbols lead to invalid results with no valid…

• NS
  National Science Foundation

  Awards: 2037745, DMR-1928882
• UD
  U.S. Department of Energy
• AW
  Amazon Web Services
• DF
  Deutsche Forschungsgemeinschaft

  Award: NFDI4-1
• SN
  Schweizerischer Nationalfonds zur Förderung der Wissenschaftlichen Forschung

  Award: 191127
• FD
  Fundação de Amparo à Pesquisa do Estado de São Paulo

  Awards: 2021/, 2021/01633-3
• CD
  Coordenação de Aperfeiçoamento de Pessoal de Nível Superior
• AS
  Austrian Science Fund

  Award: J4309
• UW
  Universität Wien
• CN
  Consejo Nacional de Ciencia y Tecnología

  Award: CVU 105568
• UO
  University of Toronto
• NC
  National Centre of Competence in Research Robotics

  Award: P2ELP2_195155
• NI
  National Institutes of Health

  Award: R35GM137966
• OO
  Office of Science
• H2
  Horizon 2020 Framework Programme
• NS
  Natural Sciences and Engineering Research Council of Canada
• ER
  European Research Council
• H2
  Horizon 2020

  Award: 666983
• NI
  National Institute of General Medical Sciences