Journal of Computational Chemistry

Alvarez‐Ginarte, Yoanna María; Marrero‐Ponce, Yovani; Ruiz‐García, José Alberto; Montero‐Cabrera, Luis Alberto; García De La Vega, Jose Manuel; Noheda Marin, Pedro; Crespo‐Otero, Rachel; Zaragoza, Francisco Torrens; García‐Domenech, Ramón

doi: 10.1002/jcc.20745pmid: 17639502

The great cost associated with the development of new anabolic–androgenic steroid (AASs) makes necessary the development of computational methods that shorten the drug discovery pipeline. Toward this end, quantum, and physicochemical molecular descriptors, plus linear discriminant analysis (LDA) were used to analyze the anabolic/androgenic activity of structurally diverse steroids and to discover novel AASs, as well as also to give a structural interpretation of their anabolic–androgenic ratio (AAR). The obtained models are able to correctly classify 91.67% (86.27%) of the AASs in the training (test) sets, respectively. The results of predictions on the 10% full‐out cross‐validation test also evidence the robustness of the obtained model. Moreover, these classification functions are applied to an “in house” library of chemicals, to find novel AASs. Two new AASs are synthesized and tested for in vivo activity. Although both AASs are less active than some commercially AASs, this result leaves a door open to a virtual variational study of the structure of the two compounds, to improve their biological activity. The LDA‐assisted QSAR models presented here, could significantly reduce the number of synthesized and tested AASs, as well as could increase the chance of finding new chemical entities with higher AAR. © 2007 Wiley Periodicals, Inc. J Comput Chem, 2008

Yasuda, Koji

doi: 10.1002/jcc.20779pmid: 17614340

We propose the algorithm to evaluate the Coulomb potential in the ab initio density functional calculation on the graphics processor unit (GPU). The numerical accuracy required for the algorithm is investigated in detail. It is shown that GPU, which supports only the single‐precision floating number natively, can take part in the major computational tasks. Because of the limited size of the working memory, the Gauss‐Rys quadrature to evaluate the electron repulsion integrals (ERIs) is investigated in detail. The error analysis of the quadrature is performed. New interpolation formula of the roots and weights is presented, which is suitable for the processor of the single‐instruction multiple‐data type. It is proposed to calculate only small ERIs on GPU. ERIs can be classified efficiently with the upper‐bound formula. The algorithm is implemented on NVIDIA GeForce 8800 GTX and the Gaussian 03 program suite. It is applied to the test molecules Taxol and Valinomycin. The total energies calculated are essentially the same as the reference ones. The preliminary results show the considerable speedup over the commodity microprocessor. © 2007 Wiley Periodicals, Inc. J Comput Chem, 2008

Stashans, Arvids; Chamba, Gaston; Pinto, Henry

doi: 10.1002/jcc.20777pmid: 17654648

The electronic structure, chemical bonding, geometry, and effects produced by Sr‐doping in CaCO3 have been studied on the basis of density‐functional theory using the VASP simulation package and molecular‐orbital theory utilizing the CLUSTERD computer code. Two calcium carbonate structures which occur naturally in anhydrous crystalline forms, calcite and aragonite, were considered in the present investigation. The obtained diagrams of density of states show similar patterns for both materials. The spatial structures are computed and analyzed in comparison to the available experimental data. The electronic properties and atomic displacements because of the trace element Sr‐incorporation are discussed in a comparative manner for the two crystalline structures. © 2007 Wiley Periodicals, Inc. J Comput Chem, 2008

Wu, Ruibo; Cao, Zexing

doi: 10.1002/jcc.20793pmid: 17591721

The combined density functional quantum mechanical/molecular mechanical (QM/MM) approach has been used to investigate methyl‐transfer reactions catalyzed by the N5‐glutamine S‐adenosyl‐L‐methionine (SAM)‐dependent methyltransferase (HemK) and the coenzyme‐modified HemK with the replacement of SAM by a nitrogen analogue. Calculations reveal that the catalytic methyl transfer by HemK is an energy‐favored process with an activation barrier of 15.7 kcal/mol and an exothermicity of 12.0 kcal/mol, while the coenzyme‐modified HemK is unable to catalyze the methyl transfer because of a substantial barrier of 20.6 kcal/mol and instability of the product intermediate. The results lend support to the experimental proposal that the nitrogen analogue of the SAM coenzyme should be a practicable inhibitor for the catalytic methyl transfer by HemK. Comparative QM/MM calculations show that the protein environment, especially the residues Asn197 and Pro198 in the active site, plays a pivotal role in stabilizing the transition state and regulating the positioning of reactive groups. © 2007 Wiley Periodicals, Inc. J Comput Chem, 2008

Fias, Stijn; Van Damme, Sofie; Bultinck, Patrick

doi: 10.1002/jcc.20794pmid: 17631652

The aromaticity and local‐aromaticity of a large set of polycyclic aromatic hydrocarbons (PAHs) is studied using multicenter delocalization indices from generalized population analysis and the popular nucleus independent chemical shift (NICS) index. A method for the fast computation of the NICS values is introduced, using the so‐called pseudo‐π‐method. A detailed examination is made of the multidimensional nature of aromaticity. The lack of a good correlation between the NICS and the multicenter delocalization indices is reported and the grounds discussed. It is shown through a thorough statistical analysis that the NICS values arise not only from local aromaticity of the benzenoid rings, but also from other circuits. It is shown that the NICS indices do not reveal the individual aromatic nature of a specific ring, contrary to the delocalization indices. © 2007 Wiley Periodicals, Inc. J Comput Chem, 2008

Castillo, Norberto; Robertson, Katherine N.; Choi, S. C.; Boyd, Russell J.; Knop, Osvald

doi: 10.1002/jcc.20795pmid: 17607718

The aim was to investigate the relationship between the bond length and the electron density at the bond critical point in homonuclear XX and ZZ and heteronuclear CZ bonds (X = Li‐F, Z = Na‐Cl). The d,ρc pairs were obtained from 472 target bonds in DFT‐optimized (B3LYP/6‐311+G(d,p)) small molecular species. These species were selected arbitrarily but with a view to maximize the range widths WR for each atom combination. It was found that (i) with one clear exception, the d(A − A) means (A = X or Z) correlate linearly with the bond lengths d(A2) of the respective diatomic molecules; (ii) the d(A − A) means correlate parabolically with n, the formal number of valence electrons in the atoms of the bond; and (iii) with increasing sample size N the ratio WR(ρc)/WR(d) appears to converge toward a representation f (WR(ρc)/WR(d))N→∞ characteristic of A. Detailed analysis of the d,ρc relationship has shown that by and large simple power regression accounts best for the DFT data. The regression coefficients of d = aρ c−b and ρc = αd−β (b, β > 0) vary with n in a seemingly irregular manner but one that is consistent with simple chemical notions. The d(A2) can be approximated in terms of multilinear MO electron occupancies. © 2007 Wiley Periodicals, Inc. J Comput Chem, 2008

Bates, P. W.; Wei, G. W.; Zhao, Shan

doi: 10.1002/jcc.20796pmid: 17591718

This article presents a novel concept, the minimal molecular surface (MMS), for the theoretical modeling of biomolecules. The MMS can be viewed as a result of the surface free energy minimization when an apolar molecule, such as protein, DNA or RNA is immersed in a polar solvent. Based on the theory of differential geometry, the MMS is created via the mean curvature minimization of molecular hypersurface functions. A detailed numerical algorithm is presented for the practical generation of MMSs. Extensive numerical experiments, including those with internal and open cavities, are carried out to demonstrated the proposed concept and algorithms. The proposed MMS is typically free of geometric singularities. Application of the MMS to the electrostatic analysis is considered for a set of twenty six proteins. © 2007 Wiley Periodicals, Inc. J Comput Chem, 2008

Norberg, Daniel; Salhi‐Benachenhou, Nessima

doi: 10.1002/jcc.20797pmid: 17607719

The stepwise and concerted pathways for the McLafferty rearrangement of the radical cations of butanal (Bu+) and 3‐fluorobutanal (3F‐Bu+) are investigated with density functional theory (DFT) and ab initio methods in conjunction with the 6‐311+G(d,p) basis set. A concerted transition structure (TS) for Bu+, (H), is located with a Gibbs barrier height of 37.7 kcal/mol as computed with CCSD(T)//BHandHLYP. Three pathways for the stepwise rearrangement of Bu+ have been located, which are all found to involve different complexes. The barrier height for the Hγ transfer is found to be 2.2 kcal/mol, while the two most favorable TSs for the Cα–Cβ cleavage are located 8.9 and 9.2 kcal/mol higher. The energies of the 3F‐Bu+ system have been calculated with the promising hybrid meta‐GGA MPWKCIS1K functional of DFT. Interestingly, the fluorine substitution yields a barrier height of only 20.5 kcal/mol for the concerted TS, (3F‐H). A smaller computed dipole moment, 12.1 D, for (3F‐H) compared with 103.2 D for (H) might explain the stabilization of the substituted TS. The Hγ transfer, with a barrier height of 4.9 kcal/mol, is found to be rate‐determining for the stepwise McLafferty rearrangement of 3F‐Bu+, in contrast to the unsubstituted case. By inspection of the spin and charge distributions of the stationary points, it is noted that the bond cleavages in the concerted rearrangements are mainly of heterolytic nature, while those in the stepwise channels are found to be homolytic. © 2007 Wiley Periodicals, Inc. J Comput Chem, 2008

Schlund, Sebastian; Müller, Robert; Graβmann, Carsten; Engels, Bernd

doi: 10.1002/jcc.20798pmid: 17631648

The determination of all possible low‐lying energy conformers of flexible molecules is of fundamental interest for various applications. It necessitates a reliable conformational search that is able to detect all important minimum structures and calculates the energies on an adequate level of theory. This work presents a strategy to identify low‐energy conformers using arginine as an example by means of a force‐field based conformational search in combination with high‐level geometry optimizations (RI‐MP2/TZVPP+). The methods used for various stages in the conformational search strategy are shown and various pitfalls are discussed. We can show that electronic energies calculated on a DFT level of theory with standard exchange‐correlation functionals strongly underestimate the intramolecular stabilization resulting from stacked orientations of the guanidine and carbonyl moiety of arginine due to the deficiency of DFT to describe dispersion effects. In this case by usage of electron correlation methods, low energy conformers comprising stacked arrangements that are counterintuitive become favorable. © 2007 Wiley Periodicals, Inc. J Comput Chem, 2008

Frison, Gilles; Ohanessian, Gilles

doi: 10.1002/jcc.20800pmid: 17631650

Although theoretical methods are now available which give very accurate results, often comparable to the experimental ones, modeling chemical or biological interesting systems often requires less demanding and less accurate theoretical methods, mainly due to computer limitations. Therefore, it is crucial to know the precision of such less reliable methods for relevant models and data. This has been done in this work for small zinc‐active site models including O‐ (H2O and OH−) and N‐donor (NH3 and imidazole) ligands. Calculations using a number of quantum mechanical methods were carried out to determine their precision for geometries, coordination number relative stability, metal–ligand bond strengths, proton affinities, and interaction energies between first and second shell ligands. We have found that obtaining chemical accuracy can be as straightforward as HF geometry optimization with a double‐ζ plus polarization basis followed by a B3LYP energy calculation with a triple‐ζ quality basis set including diffuse and polarization functions. The use of levels as low as PM3 geometry optimization followed by a B3LYP single‐point energy calculation with a double‐ζ quality basis including polarization functions already yields useful trends in bond length, proton affinities or bond dissociation energies, provided that appropriate caution is taken with the optimized structures. The reliability of these levels of calculation has been successfully demonstrated for real biomimetic cases. © 2007 Wiley Periodicals, Inc. J Comput Chem, 2008