One-dimensional molecular representations and similarity calculations: Methodology and validation

Drug discovery research is increasingly dedicated to biological screening o n a massive scale, which seems to imply a basic rejection of many computer- assisted techniques originally designed to add rationality to the early sta ges of discovery. While ever-faster and more clever 3D methodologies contin ue to be developed and rejected as alternatives to indiscriminant screening , simpler tools based on 2D structure have carved a stable niche in the hig h-throughput paradigm of drug discovery. Their staying power is due in no s mall part to simplicity, ease of use, and demonstrated ability to explain s tructure-activity data. This observation led us to wonder whether an even s impler view of structure might offer an advantage over existing 2D and 3D m ethods. Accordingly, we introduce 1D representations of chemical structure, which are generated by collapsing a 3D molecular model or a 2D chemical gr aph onto a single coordinate of atomic positions. Atoms along this coordina te are differentiated according to elemental type, hybridization, and conne ctivity. By aligning 1D representations to match up identical atom types, a measure of overall structural similarity is afforded. In extensive structu re-activity validation tests, 1D similarities consistently outperform both Daylight 2D fingerprints and Cerius(2) pharmacophore fingerprints, suggesti ng that this new, simple means of representing and comparing structures may offer a significant advantage over existing tried-and-true methods.