No abstract

To finalize a scheme intended to calculate directly approximate charge distributions for proteins, parameters for a modified PEOE method are obtained for both ionic and aromatic functional groups. For ionic groups, the electrostatic potentials of several ionic molecules (computed with the 6-3 1G** basis set) are used as constraints; for aromatic groups, the experimental gas-phase dipole and quadrupole moments of several aromatic molecules are used as constraints. The electrostatic potentials calculated with the modified PEOE point charges of ionic molecules are quite close to the ab initio electrostatic potential surfaces, and the dipole and quadrupole moments calculated with the modified PEOE charges of aromatic molecules agree well with experimental data. The method enables the charge distribution, described as a set of atom-centered monopoles, to be calculated directly. This avoids assumptions concerning transferability of charges from small model compounds and describes in an approximate fashion the redistribution of charge upon assembly of amino acid residues into polypeptide molecules. The computational requirements for the procedure are negligible. (12) Toluene: Rudolph, H. D.; Dreizler, H.; Jaeschke, A.; Wendling, P. 2. Naturforsch. 1967, 220, 940. o-Xylene: Rudolph, H. D.; Walzer, K.; Krutzik, I. J. Mol. Spectrom. 1973, 47, 314. Phenol: Pedersen, T.; Larsen, N. W.; Nygaard, L. J. Mol. S r r c t . 1969.4, 59. Aniline: Lister, D. G.; Tyler, J. K.; Hog, J. N.; Larsen, N. W. J. Mol. Struct. 1974, 23, 253. Pyridine: Sorensen, G . 0.; Mahler, L.; RastrupAndersen, N. J. Mol. Struct. 1974, 20, 119. 2-Methylpyridine: Dreizler, H.; Rudolph, H. D.; Miider, H. 2. Na-turJorsch. 1970, 25a, 25. 4-Methylpyridine: Rudolph, H. D.; Dreizler, H.; Seiler, H. Z . Naturforsch. 1967, 22a, 1738. Pyrrole: Nygaard, L.; Nielsen, J. T.; Kirchheiner, J.; Maltesen, G.; RastrupAndersen, J.; Ssrensen, G. 0. J. Mol. Slruct. 1%9, 3, 491. N-methylpyrrole: Arnold, W.; Dreizler, H.; Rudolph, H. D.

To investigate the stability of ionic pairs in aqueous solution, polarizable continuum-model (PCM) ab initio molecular orbital calculations were carried out. Ion pairs formed from three organic ions, methylacetate, methylammonium, and guanidinium, were treated. The methylacetate ion pair forms a stable complex when the two methyl groups are in contact. When the charged groups of methylacetate and methylammonium are in contact, this pair does not form a stable complex in aqueous solution. When the methyl group of methylacetate is in contact with that of methylammonium, the ionic groups are far apart, and a stable complex is not formed, even though the complex involves a strong attractive electrostatic interaction. The guanidinium pair forms a stable complex in aqueous solution at all conformations investigated in this work even though a strong repulsive electrostatic interaction is present in this complex.

Summary Early quantitative structure-activity relationship (QSAR) technologies have unsatisfactory versatility and accuracy in fields such as drug discovery because they are based on traditional machine learning and interpretive expert features. The development of Big Data and deep learning technologies significantly improve the processing of unstructured data and unleash the great potential of QSAR. Here we discuss the integration of wet experiments (which provide experimental data and reliable verification), molecular dynamics simulation (which provides mechanistic interpretation at the atomic/molecular levels), and machine learning (including deep learning) techniques to improve QSAR models. We first review the history of traditional QSAR and point out its problems. We then propose a better QSAR model characterized by a new iterative framework to integrate machine learning with disparate data input. Finally, we discuss the application of QSAR and machine learning to many practical research fields, including drug development and clinical trials.

No abstract