An approach based on genetic algorithms and machine learning coupled for studying alloy and molecular clusters by optimizing quantum energy surfaces.

A new genetic algorithm has been proposed focusing on direct ab initio potential energy surface (PES) global minima search. Besides the commonly used operators, this new approach uses an operator to: improve the initial cluster generation, classify and compare all generated clusters, and use machine learning to model the quantum PES used in parallel optimization. Part of the validation process for this methodology was done with C u n A u m ( n + m ≤ X for X = 14 , 19 , 38 , 55 ) and A u n A g n ( n = 10 , 20 , 30 , 40 , 50 , 60 , 70 , and 75). The results are in fair agreement with the literature and led to a new global minimum for C u 12 A u 7 . A search has been done for the lowest energies of L i n nanoclusters with 2-8 atoms using the DFT approach and for L i 3 , L i 4 , L i 2 H , L i 3 H using DLPNO-CCSD(T) approach. NQGA successfully performed the MP2 optimizations for ( H 2 O ) 11 cluster. In all cases, the proposed genetic algorithm located the previously reported global minima with very efficient performance. The new proposed methodology makes it possible to optimize cluster geometries directly using high-level ab initio methods relinquishing any bias introduced by a classical approach. Our results show that this proposed method has great potential applications due to its flexibility and efficiency in identifying global minima in the tested atomic systems.

Application of a quantum genetic algorithm and QTAIM analysis in the study of structural and electronic properties of neutral bimetallic clusters NaxLiy (4 ≤ x + y ≤ 10).

Alloy clusters of NaxLiy (4 ≤ x + y ≤ 10) are studied by exploring the potential energy surface in the ab initio MP2 level with the support of a quantum genetic algorithm (QGA). In some cases, the structures have been also refined with DFT and coupled-cluster methods. The general trends of sodium-lithium structures are in line with previous studies. The ionization potentials and polarizabilities to all structures were calculated with MP2 method and the average error between these two properties compared with experimental data was 6% and 13%, respectively. The topological analysis based on quantum theory of atoms in molecules (QTAIM) showed that by increasing the cluster size of the diatomic system there was a decrease of atomic interaction energies. The degree of degeneracy from D3BIA aromaticity index and the analysis of the atomic charges showed the influence (by charge transfer) of the chemical element in lower quantity in the cluster with respect to the other atoms. Our achievements of comparing our theoretical results with available experimental data have demonstrated that our approach can also predict satisfactorily quantum atomic and alloy clusters properties, at least, for low nuclearities.

1H and 195Pt NMR prediction for inclusion compounds formed by cisplatin and oxidized carbon nanostructures.

Prediction of NMR chemical shifts can assist experimentalists in the characterization of drug delivery systems based on carbon nanocomposites. Chemical shifts are strongly correlated to the nucleus position and its chemical neighborhood. Therefore, to predict structures and NMR properties of complex chemical models, choosing a more consistent theoretical level capable of providing more realistic results and moderate computational demand is a major challenge. In this work, we predicted the NMR spectra of inclusion compounds formed by cisplatin (cDDP) and an oxidized carbon nanotube (CNTox) and nanocone (CNCox) considered by specialists as potential drug delivery systems. The 195Pt NMR chemical shifts calculated at the DFT level with the new relativistic NMR-DKH basis set were -2314 ppm and -2192 ppm for cDDP@CNTox and cDDP@CNCox complexes, respectively, which are both high-field shifted relative to the free cDDP (-2110 ppm). 1H NMR chemical shifts are also sensitive to the inclusion process. The H (NH3) signals are found on average at +4.3 (cDDP), -5.1 (cDDP@CNTox) and +6.6 ppm (cDDP@CNCox). Interestingly, despite the similar inclusion modes in CNTox and CNCox cavities, the 1H NMR shifts were in opposite directions. A possible reason might be the higher stability of cDDP@CNTox (ΔE F = -19.9 kcal mol-1) than that of cDDP@CNCox (ΔE F = -5.7 kcal mol-1), which suggests a short guest-host contact in the former and consequently, a more efficient shielding of hydrogen atoms due to the electron-rich carbon structure. These results may be helpful as comparison data in the NMR spectra assignment in solution and the inclusion compounds' structural elucidation.

A New Genetic Algorithm Approach Applied to Atomic and Molecular Cluster Studies.

A new procedure is suggested to improve genetic algorithms for the prediction of structures of nanoparticles. The strategy focuses on managing the creation of new individuals by evaluating the efficiency of operators (o 1, o 2,…,o 13) in generating well-adapted offspring. This is done by increasing the creation rate of operators with better performance and decreasing that rate for the ones which poorly fulfill the task of creating favorable new generation. Additionally, several strategies (thirteen at this level of approach) from different optimization techniques were implemented on the actual genetic algorithm. Trials were performed on the general case studies of 26 and 55-atom clusters with binding energy governed by a Lennard-Jones empirical potential with all individuals being created by each of the particular thirteen operators tested. A 18-atom carbon cluster and some polynitrogen systems were also studied within REBO potential and quantum approaches, respectively. Results show that our management strategy could avoid bad operators, keeping the overall method performance with great confidence. Moreover, amongst the operators taken from the literature and tested herein, the genetic algorithm was faster when the generation of new individuals was carried out by the twist operator, even when compared to commonly used operators such as Deaven and Ho cut-and-splice crossover. Operators typically designed for basin-hopping methodology also performed well on the proposed genetic algorithm scheme.

Theoretical and Experimental Studies of the Controlled Release of Tetracycline Incorporated into Bioactive Glasses.

Several authors have studied the release profile of drugs incorporated in different devices. However, to the best of our knowledge, although many studies have been done on the release of tetracycline, in these release devices, no study has investigated if the released compound is actually the tetracycline, or, instead, a degraded product. This approach is exploited here. In this work, we analyse the influence of two drying methods on the tetracycline delivery behaviour of synthesised glasses using the sol-gel process. We compare the drying methods results using both theoretical models and practical essays, and analyse the chemical characteristic of the released product in order to verify if it remains tetracycline. Samples were freeze-dried or dried in an oven at 37°C and characterised by several methods such as Fourier transform infrared spectroscopy (FTIR), scanning electron microscopy (SEM), thermogravimetric analysis (TG), differential thermogravimetric analysis (DTG), differential thermal analyses (DTA) and gas adsorption analysis (BET). The released concentration of tetracycline hydrochloride was studied as a function of time, and it was measured by ultraviolet spectrophotometry in the tetracycline wavelength. The drug delivery profiles were reasonably consistent with a diffusion model analysis. In addition, we observed higher release rates for the freeze-dried compared to those dried in an oven at 37°C. This higher release can be attributed to larger pore size for the freeze-dried sample systems with tetracycline, which promoted more water penetration, improving the drug diffusion. The analysis of the solution obtained in the release tests using high-performance liquid chromatography- mass spectrometry (HPLC-MS) confirmed that tetracycline was being released.

Structure and stability of neutral Al-Mg nanoclusters up to 55 atoms.

The geometries of aluminum-magnesium nanoalloys are explored using a genetic algorithm applied to the Gupta potential function and tuned to search for the 10 lowest energy minima for each cluster size and composition. Each structure is re-optimized using density functional theory calculations, allowing both a classical and quantum analysis of the system. Average binding energies, excess energies and HOMO-LUMO gaps are calculated. Differences between classical and quantum descriptions are discussed. The results reveal structures that are particularly stable, and a discussion on magic numbers of valence electrons is presented.

Genetic algorithms coupled with quantum mechanics for refinement of force fields for RNA simulation: a case study of glycosidic torsions in the canonical ribonucleosides.

We report the use of genetic algorithms (GA) as a method to refine force field parameters in order to determine RNA energy. Quantum-mechanical (QM) calculations are carried out for the isolated canonical ribonucleosides (adenosine, guanosine, cytidine and uridine) that are taken as reference data. In this particular study, the dihedral and electrostatic energies are reparametrized in order to test the proposed approach, i.e., GA coupled with QM calculations. Overall, RMSE comparison with recent published results for ribonucleosides energies shows an improvement, on average, of 50%. Finally, the new reparametrized potential energy function is used to determine the spatial structure of RNA (PDB code ) that was not taken into account in the parametrization process. This structure was improved about 82% comparable with previously published results.

Theoretical study of small sodium-potassium alloy clusters through genetic algorithm and quantum chemical calculations.

Genetic algorithm is employed to survey an empirical potential energy surface for small Na(x)K(y) clusters with x + y ≤ 15, providing initial conditions for electronic structure methods. The minima of such empirical potential are assessed and corrected using high level ab initio methods such as CCSD(T), CR-CCSD(T)-L and MP2, and benchmark results are obtained for specific cases. The results are the first calculations for such small alloy clusters and may serve as a reference for further studies. The validity and choice of a proper functional and basis set for DFT calculations are then explored using the benchmark data, where it was found that the usual DFT approach may fail to provide the correct qualitative result for specific systems. The best general agreement to the benchmark calculations is achieved with def2-TZVPP basis set with SVWN5 functional, although the LANL2DZ basis set (with effective core potential) and SVWN5 functional provided the most cost-effective results.

Analysis of the applicability of artificial neural networks for studying blood plasma: determination of magnesium ion concentration as a case study.

Artificial neural networks are suggested for use in predicting metal ion concentration in human blood plasma. Simulated and available experimental data are used to train the artificial neural network. Particularly, using 850 simulated samples, the network predicted the magnesium-free ion concentration with an average error smaller than 1%. Clinical data recently reported for 20 patients were considered and the artificial neural network predicted the concentration of free magnesium ion with an average error of about 6%. Overall, the approach of using artificial neural networks as an alternative or complementary strategy to deal with the analysis of human blood plasma can be useful for clinical diagnostics, if there is sufficient data to train the artificial neural network.

An alternative approach based on artificial neural networks to study controlled drug release.

An alternative methodology based on artificial neural networks is proposed to be a complementary tool to other conventional methods to study controlled drug release. Two systems are used to test the approach; namely, hydrocortisone in a biodegradable matrix and rhodium (II) butyrate complexes in a bioceramic matrix. Two well-established mathematical models are used to simulate different release profiles as a function of fundamental properties; namely, diffusion coefficient (D), saturation solubility (C(s)), drug loading (A), and the height of the device (h). The models were tested, and the results show that these fundamental properties can be predicted after learning the experimental or model data for controlled drug release systems. The neural network results obtained after the learning stage can be considered to quantitatively predict ideal experimental conditions. Overall, the proposed methodology was shown to be efficient for ideal experiments, with a relative average error of <1% in both tests. This approach can be useful for the experimental analysis to simulate and design efficient controlled drug-release systems.