NMRmix: A Tool for the Optimization of Compound Mixtures in 1D ¹H NMR Ligand Affinity Screens

Abstract

NMR ligand affinity screening is a powerful technique that is routinely used in drug discovery or functional genomics to directly detect protein–ligand binding events. Binding events can be identified by monitoring differences in the 1D 1H NMR spectrum of a compound with and without protein. Although a single NMR spectrum can be collected within a short period (2—10 min per sample), one-by-one screening of a protein against a library of hundreds or thousands of compounds requires a large amount of spectrometer time and a large quantity of protein. Therefore, compounds are usually evaluated in mixtures ranging in size from 3 to 20 compounds to improve the efficiency of these screens in both time and material. Ideally, the NMR signals from individual compounds in the mixture should not overlap so that spectral changes can be associated with a particular compound. We have developed a software tool, NMRmix, to assist in creating ideal mixtures from a large panel of compounds with known chemical shifts. Input to NMRmix consists of an 1H NMR peak list for each compound, a user-defined overlap threshold, and additional user-defined parameters if default settings are not used. NMRmix utilizes a simulated annealing algorithm to optimize the composition of the mixtures to minimize spectral peak overlaps so that each compound in the mixture is represented by a maximum number of nonoverlapping chemical shifts. A built-in graphical user interface simplifies data import and visual evaluation of the results.

Conserved lipid and small molecule modulation of COQ8 reveals regulation of the ancient UbiB family

“…To minimize the materials required to evaluate every compound with COQ8A ND250 , the compounds were combined into 89 mixtures of 3-5 compounds each. The mixtures were designed to minimize 1 H peak signal overlap between the compounds using NMRmix (Stark et al, 2016). and NMRbot were used to automate the data collection process (Clos et al, 2013).…”

Current NMR Techniques for Structure-Based Drug Discovery

“…In general, experimental designs aim at efficient, prompt, and complete screening, with thousands of 19 F-chemicals being divided into mixtures, typically consisting of 10–20 fragments, and hit screening is sequentially executed for each mixture to improve its throughput [ 62 ]. Hit candidates are identified by 19 F-NMR signal changes from 10 to 20 merged peaks.…”

“…All compounds in the 19 F-chemical library should contain fluorine atoms, which may limit the library size and structural variety of the compounds. Assessment of the quality control of the chemicals, prior to initiating the NMR-based screening experiments, significantly influences the results [ 16 , 62 ]. Stock solution of the chemicals, typically dissolved in dimethyl sulfoxide (DMSO- d 6 ) and the concentration of fragments is adjusted to 40–100 mM, is diluted with aqueous buffer to 40–100 μM.…”

Review of emerging metabolomic tools and resources: 2015–2016

“…NMRmix is a tool that supports creation of ideal mixtures from a large panel of compounds with known chemical shifts wherein it uses a simulated annealing algorithm to optimize the composition of the mixtures to minimize spectral peak overlaps, and a graphical user interface simplifies data import and visualization .…”