Capturing Mixtures ” Bringing Informatics to the World of Practical Chemistry”, Recorded live December 19, 2019

Our industry has been using cheminformatics to support drug discovery for decades, leveraging formats for describing organic molecules, such as Molfile, SMILES, and InChI. These are idealized concepts rather than a description of the laboratory reality: it is rare that a substance can be accurately described with a single molecule.

Almost every sample has an impurity level, or is dissolved in solvent, or exists as an adduct, or is explicitly combined with other substances. Mixtures often combine certainty and uncertainty within the same description: some components can be well-defined molecules with an accurately measured molar concentration, while others are estimated portions, or amorphous adjuncts.

The cheminformatics community has yet to select a standard format for describing mixtures in a machine-readable, standardized, and interoperable way, and most publications fall back to using a text description. Electronic lab notebooks and inventory databases are forced to choose between using text, proprietary formats, or ignoring mixture composition altogether.

We will describe our work toward two new data structures: Mixfile and MInChI, which are intended to fill roles that are analogous to Molfile and InChI, respectively. We will describe the ways in which we expect that mixtures-based informatics tools will affect all industries that intersect with chemistry.

Join our expert panelists as we discuss the future of mixture representation, including:

• Introduction of the new mixtures standards, MInChI and Mixfile
• Impacts on health and chemical safety communities
• New technologies for unambiguous mixture capture