Alternative CNDOL Fockians for fast and accurate description of molecular exciton properties.

CNDOL is an a priori, approximate Fockian for molecular wave functions. In this study, we employ several modes of singly excited configuration interaction (CIS) to model molecular excitation properties by using four combinations of the one electron operator terms. Those options are compared to the experimental and theoretical data for a carefully selected set of molecules. The resulting excitons are represented by CIS wave functions that encompass all valence electrons in the system for each excited state energy. The Coulomb-exchange term associated to the calculated excitation energies is rationalized to evaluate theoretical exciton binding energies. This property is shown to be useful for discriminating the charge donation ability of molecular and supermolecular systems. Multielectronic 3D maps of exciton formal charges are showcased, demonstrating the applicability of these approximate wave functions for modeling properties of large molecules and clusters at nanoscales. This modeling proves useful in designing molecular photovoltaic devices. Our methodology holds potential applications in systematic evaluations of such systems and the development of fundamental artificial intelligence databases for predicting related properties.

Electronic Descriptors for Supervised Spectroscopic Predictions.

Spectroscopic properties of molecules hold great importance for the description of the molecular response under the effect of UV/vis electromagnetic radiation. Computationally expensive ab initio (e.g., MultiConfigurational SCF, Coupled Cluster) or TDDFT methods are commonly used by the quantum chemistry community to compute these properties. In this work, we propose a (supervised) Machine Learning approach to model the absorption spectra of organic molecules. Several supervised ML methods have been tested such as Kernel Ridge Regression (KRR), Multiperceptron Neural Networs (MLP), and Convolutional Neural Networks. [Ramakrishnan et al. J. Chem. Phys. 2015, 143, 084111. Ghosh et al. Adv. Sci. 2019, 6, 1801367.] The use of only geometrical-atomic number descriptors (e.g., Coulomb Matrix) proved to be insufficient for an accurate training. [Ramakrishnan et al. J. Chem. Phys. 2015, 143, 084111.] Inspired by the TDDFT theory, we propose to use a set of electronic descriptors obtained from low-cost DFT methods: orbital energy differences (Δϵia = ϵa - ϵi), transition dipole moment between occupied and unoccupied Kohn-Sham orbitals (⟨ϕi|r|ϕa⟩), and when relevant, charge-transfer character of monoexcitations (Ria). We demonstrate that with these electronic descriptors and the use of Neural Networks we can predict not only a density of excited states but also get a very good estimation of the absorption spectrum and charge-transfer character of the electronic excited states, reaching results close to chemical accuracy (∼2 kcal/mol or ∼0.1 eV).

1,3 Dipolar Cycloaddition of Münchnones: Factors behind the Regioselectivity.

The 1,3 dipolar cycloaddition reactions of münchnones and alkenes provide an expedite synthetic way to substituted pyrroles, an exceedingly important structural motif in the pharmaceutical and material science fields of research. The factors governing their regioselectivity rationalization are not well understood. Using several approaches, we investigate a set of 14 reactions (featuring two münchnones, 12 different alkenes, and two alkynes). The Natural Bond Theory and the Non-Covalent Interaction Index analyses of the noncovalent interaction energies fail to predict the experimental major regioisomer. Employing global cDFT descriptors or local ones such as the Fukui function and dual descriptor yields similarly inaccurate predictions. Only the local softness pairing, within Pearson's Hard and Soft Acids and Bases principle, constitutes a reliable predictor for the major reaction product. By taking into account an estimator for the steric effects, the correct regioisomer is predicted. Steric effects play a major role in driving the regioselectivity, as was corroborated by energy decomposition analysis of the transition states.

MDSCAN: RMSD-based HDBSCAN clustering of long molecular dynamics.

MOTIVATION: The term clustering designates a comprehensive family of unsupervised learning methods allowing to group similar elements into sets called clusters. Geometrical clustering of molecular dynamics (MD) trajectories is a well-established analysis to gain insights into the conformational behavior of simulated systems. However, popular variants collapse when processing relatively long trajectories because of their quadratic memory or time complexity. From the arsenal of clustering algorithms, HDBSCAN stands out as a hierarchical density-based alternative that provides robust differentiation of intimately related elements from noise data. Although a very efficient implementation of this algorithm is available for programming-skilled users (HDBSCAN*), it cannot treat long trajectories under the de facto molecular similarity metric RMSD. RESULTS: Here, we propose MDSCAN, an HDBSCAN-inspired software specifically conceived for non-programmers users to perform memory-efficient RMSD-based clustering of long MD trajectories. Methodological improvements over the original version include the encoding of trajectories as a particular class of vantage-point tree (decreasing time complexity), and a dual-heap approach to construct a quasi-minimum spanning tree (reducing memory complexity). MDSCAN was able to process a trajectory of 1 million frames using the RMSD metric in about 21 h with <8 GB of RAM, a task that would have taken a similar time but more than 32 TB of RAM with the accelerated HDBSCAN* implementation generally used. AVAILABILITY AND IMPLEMENTATION: The source code and documentation of MDSCAN are free and publicly available on GitHub (https://github.com/LQCT/MDScan.git) and as a PyPI package (https://pypi.org/project/mdscan/). SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

RCDPeaks: memory-efficient density peaks clustering of long molecular dynamics.

MOTIVATION: Density Peaks is a widely spread clustering algorithm that has been previously applied to Molecular Dynamics (MD) simulations. Its conception of cluster centers as elements displaying both a high density of neighbors and a large distance to other elements of high density, particularly fits the nature of a geometrical converged MD simulation. Despite its theoretical convenience, implementations of Density Peaks carry a quadratic memory complexity that only permits the analysis of relatively short trajectories. RESULTS: Here, we describe DP+, an exact novel implementation of Density Peaks that drastically reduces the RAM consumption in comparison to the scarcely available alternatives designed for MD. Based on DP+, we developed RCDPeaks, a refined variant of the original Density Peaks algorithm. Through the use of DP+, RCDPeaks was able to cluster a one-million frames trajectory using less than 4.5 GB of RAM, a task that would have taken more than 2 TB and about 3× more time with the fastest and less memory-hunger alternative currently available. Other key features of RCDPeaks include the automatic selection of parameters, the screening of center candidates and the geometrical refining of returned clusters. AVAILABILITY AND IMPLEMENTATION: The source code and documentation of RCDPeaks are free and publicly available on GitHub (https://github.com/LQCT/RCDPeaks.git). SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

BitQT: a graph-based approach to the quality threshold clustering of molecular dynamics.

MOTIVATION: Classical Molecular Dynamics (MD) is a standard computational approach to model time-dependent processes at the atomic level. The inherent sparsity of increasingly huge generated trajectories demands clustering algorithms to reduce other post-simulation analysis complexity. The Quality Threshold (QT) variant is an appealing one from the vast number of available clustering methods. It guarantees that all members of a particular cluster will maintain a collective similarity established by a user-defined threshold. Unfortunately, its high computational cost for processing big data limits its application in the molecular simulation field. RESULTS: In this work, we propose a methodological parallel between QT clustering and another well-known algorithm in the field of Graph Theory, the Maximum Clique Problem. Molecular trajectories are represented as graphs whose nodes designate conformations, while unweighted edges indicate mutual similarity between nodes. The use of a binary-encoded RMSD matrix coupled to the exploitation of bitwise operations to extract clusters significantly contributes to reaching a very affordable algorithm compared to the few implementations of QT for MD available in the literature. Our alternative provides results in good agreement with the exact one while strictly preserving the collective similarity of clusters. AVAILABILITY AND IMPLEMENTATION: The source code and documentation of BitQT are free and publicly available on GitHub (https://github.com/LQCT/BitQT.git) and ReadTheDocs (https://bitqt.readthedocs.io/en/latest/), respectively. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

MCSS-Based Predictions of Binding Mode and Selectivity of Nucleotide Ligands.

Computational fragment-based approaches are widely used in drug design and discovery. One of their limitations is the lack of performance of docking methods, mainly the scoring functions. With the emergence of fragment-based approaches for single-stranded RNA ligands, we analyze the performance in docking and screening powers of an MCSS-based approach. The performance is evaluated on a benchmark of protein-nucleotide complexes where the four RNA residues are used as fragments. The screening power can be considered the major limiting factor for the fragment-based modeling or design of sequence-selective oligonucleotides. We show that the MCSS sampling is efficient even for such large and flexible fragments. Hybrid solvent models based on some partial explicit representations improve both the docking and screening powers. Clustering of the n best-ranked poses can also contribute to a lesser extent to better performance. A detailed analysis of molecular features suggests various ways to optimize the performance further.

In silico strategy for detailing the binding modes of a novel family of peptides proven as ghrelin receptor agonists.

Ghrelin is a peptide hormone involved in multiple functions, including growth hormone release stimulation, food intake regulation, and metabolic and cytoprotective effect. A novel family of peptides with internal cycles was designed as ghrelin analogs and the biological activity of two of them (A228 and A233) was experimentally studied in-depth. In this work, an in silico strategy was developed for describing and assessing the binding modes of A228 and A233 to GHS-R1a (ghrelin receptor) comparing it with ghrelin and GHRP-6 peptides. Several reported structures of different G protein coupled receptors were used as templates, to obtain a good quality model of GHS-R1a. The best model was selected by preliminary molecular docking with ghrelin and GHRP-6. Docking was used to estimate peptide orientations in the binding site of the best model, observing a superposition of its N-terminal and its first aromatic residue. To test the complex stability in time, the C-terminal fragments of each peptide were added and the complexes were inserted a 1-palmitoyl-2-oleoyl-sn-glycero-3-phosphocholine (POPC) membrane, performing a molecular dynamic simulation for 100 ns using the CHARMM36 force field. Despite of the structural differences, the studied peptides share a common binding mode; the N-terminal interacts with E124 and the aromatic residue close to it, with the aromatic cluster (F279, F309, and F312). A preliminary pharmacophore model, consisting in a positive charged amine and an aromatic ring at an approximate distance of 0.79 nm, can be proposed. The results here described could represent a step forward in the efficient search of new ghrelin analogs.

A computational strategy to understand structure-activity relationship of 1,3-disubstituted imidazole [1,5-α] pyrazine derivatives described as ATP competitive inhibitors of the IGF-1 receptor related to Ewing sarcoma.

We followed a comprehensive computational strategy to understand and eventually predict the structure-activity relationship of thirty-three 1,3-disubstituted imidazole [1,5-α] pyrazine derivatives described as ATP competitive inhibitors of the IGF-1 receptor related to Ewing sarcoma. The quantitative structure-activity relationship model showed that the inhibitory potency is correlated with the molar volume, a steric descriptor and the net charge calculated value on atom C1 (q1) and N4 (q4) of the pharmacophore, all of them appearing to give a positive contribution to the inhibitory activity. According to experimental and calculated values, the most potent compound would be 3-[4-(azetidin-2-ylmethyl) cyclohexyl]-1-[3-(benzyloxy) phenyl] imidazo [1,5-α]pyrazin-8-amine (compound 23). Docking was used to guess important residues involved in the ATP-competitive inhibitory activity. It was validated by 200 ns of molecular dynamics (MD) simulation using improved linear interaction energy (LIE) method. MD of previously preferred structures by docking shows that the most potent ligand could establish hydrogen bonds with the ATP-binding site of the receptor, and the Ser979 and Ser1059 residues contribute favourably to the binding stability of compound 23. MD simulation also gave arguments about the chemical structure of the compound 23 being able to fit in the ATP-binding pocket, expecting to remain stable into it during the entire simulation and allowing us to hint the significant contribution expected to be given by electrostatic and hydrophobic interactions to the ligand-receptor complex stability. This computational combined strategy here described could represent a useful and effective prime approach to guide the identification of tyrosine kinase inhibitors as new lead compounds.

Understanding the disrupting mechanism of the Tau aggregation motif "306 VQIVYK311 " by phenylthiazolyl-hydrazides inhibitors.

Alzheimer's disease is a progressive neurodegenerative disorder characterized by the abnormal processing of the Tau and the amyloid precursor proteins. The unusual aggregation of Tau is based on the formation of intermolecular ß-sheets through two motifs: 275 VQIINK280 and 306 VQIVYK311 . Phenylthiazolyl-hydrazides (PTHs) are capable of inhibiting/disassembling Tau aggregates. However, the disaggregation mechanism of Tau oligomers by PTHs is still unknown. In this work, we studied the disruption of the oligomeric form of the Tau motif 306 VQIVYK311 by PTHs through molecular docking, molecular dynamics, and free energy calculations. We predicted hydrophobic interactions as the major driving forces for the stabilization of Tau oligomer, with V306 and I308 being the major contributors. Nonpolar component of the binding free energy is essential to stabilize Tau-PTH complexes. PTHs disrupted mainly the van der Waals interactions between the monomers, leading to oligomer destabilization. Destabilization of full Tau filament by PTHs and emodin was not observed in the sampled 20 ns; however, in all cases, the nonpolar component of the binding free energy is essential for the formation of Tau filament-PTH and Tau filament-emodin. These results provide useful clues for the design of more effective Tau-aggregation inhibitors.

Theoretical Evaluation of the Molecular Inclusion Process between Chlordecone and Cyclodextrins: A New Method for Mitigating the Basis Set Superposition Error in the Case of an Implicit Solvation Model.

The aim of this work is to describe the molecular inclusion of chlordecone with α-, ß-, and γ-cyclodextrin in aqueous solution using quantum mechanics. The guest-host complexes of chlordecone and cyclodextrins are modeled in aqueous solution using the multiple minima hypersurface methodology with a PM6-D3H4X semiempirical Hamiltonian, and the lowest energy minima obtained are reoptimized using the M06-2X density functional and the intermolecular interactions described using quantum theory of atoms in molecules (QTAIM). The studied complexes are classified according to the degree of inclusion, namely, total occlusion, partial occlusion, and external interaction. More stable complexes are obtained when γ-CD is used as the host molecule. The interactions characterized through QTAIM analysis are all of electrostatic nature, predominantly of dispersive type. In this work, a method based on the counterpoise correction is also discussed to mitigate the basis set superposition error in density functional theory calculations when using an implicit solvation model.

Role of Augmented Basis Sets and Quest for ab Initio Performance/Cost Alternative to Kohn-Sham Density Functional Theory.

Following a previous work, we have assessed the feasibility of MP2/CBS(d, t) as an alternative to state-of-the-art density functionals. The effect of using augmented basis sets is here tested on the 76 barrier heights and 10 isomerization reactions previously utilized. Moreover, calculations for 20 sets of the GMTKN24 database for thermochemistry, kinetics, and noncovalent interactions have been performed. For the density functional theory calculations, M06-2X and B3LYP-D3 functionals are utilized as two representative functionals, while MP2 and CCSD(T) methods are employed as the ab initio counterparts. The results show that MP2 calculations perform similarly to the ones obtained with M06-2X insofar as accuracy and computational cost are concerned. For all methods, the use of augmented basis sets yields enhanced results for anionic systems when compared with the ones from non-augmented bases. Otherwise, the basis-set change effect is found to be minimal. It is therefore concluded that the use of large basis sets is unjustified when facing the increase in computational cost.

Quality Threshold Clustering of Molecular Dynamics: A Word of Caution.

Clustering Molecular Dynamics trajectories is a common analysis that allows grouping together similar conformations. Several algorithms have been designed and optimized to perform this routine task, and among them, Quality Threshold stands as a very attractive option. This algorithm guarantees that in retrieved clusters no pair of frames will have a similarity value greater than a specified threshold, and hence, a set of strongly correlated frames are obtained for each cluster. In this work, it is shown that various commonly used software implementations are flawed by confusing Quality Threshold with another simplistic well-known clustering algorithm published by Daura et al. (Daura, X.; van Gunsteren, W. F.; Jaun, B.; Mark, A. E.; Gademann, K.; Seebach, D. Peptide Folding: When Simulation Meets Experiment. Angew. Chemie Int. Ed. 1999, 38 (1/2), 236-240). Daura's algorithm does not impose any quality threshold for the frames contained in retrieved clusters, bringing unrelated structural configurations altogether. The advantages of using Quality Threshold whenever possible to explore Molecular Dynamic trajectories is exemplified. An in-house implementation of the original Quality Threshold algorithm has been developed in order to illustrate our comments, and its code is freely available for further use by the scientific community.

BitClust: Fast Geometrical Clustering of Long Molecular Dynamics Simulations.

The growing computational capacity allows the investigation of large biomolecular systems by increasingly extensive molecular dynamics simulations. The resulting huge trajectories demand efficient partition methods to discern relevant structural dissimilarity. Clustering algorithms are available to address this task, but their implementations still need to be improved to gain in computational speed and to reduce the consumption of random access memory. We propose the BitClust code which, based on a combination of Python and C programming languages, performs fast structural clustering of long molecular trajectories. BitClust takes advantage of bitwise operations applied to a bit-encoded pairwise similarity matrix. Our approach allowed us to process a half-million frame trajectory in 6 h using less than 35 GB, a task that is not affordable with any of the similar alternatives.

Conceptual DFT analysis of the regioselectivity of 1,3-dipolar cycloadditions: nitrones as a case of study.

The regioselectivity of the 1,3-dipolar cycloaddition of a model nitrone with a set of dipolarophiles, presenting diverse electronic effects, is analyzed using conceptual density functional theory (DFT) methods. We deviate from standard approaches based on frontier molecular orbitals and formulations of the local hard/soft acid/base principle and use instead the dual descriptor. A detailed analysis is carried out to determine the influence of the way to calculate the dual descriptor, the computational procedure, basis set and choice of method to condensate the values of this descriptor. We show that the qualitative regioselectivity predictions depend on the choice of "computational conditions", something that indicates the danger of using black-box computational set-ups in conceptual DFT studies.

Assessing How Correlated Molecular Orbital Calculations Can Perform versus Kohn-Sham DFT: Barrier Heights/Isomerizations.

To assess the title issue, 38 hydrogen transfer barrier heights and 38 non-hydrogen transfer barrier heights/isomerizations extracted from extensive databases have been considered, in addition to 4 2 p-isomerization reactions and 6 others for large organic molecules. All Kohn-Sham DFT calculations have employed the popular M06-2X functional, whereas the correlated molecular orbital (MO)-based ones are from single-reference MP2 and CCSD(T) methods. They have all utilized the same basis sets, with raw MO energies subsequently extrapolated to the complete basis set limit without additional cost. MP2 calculations are found to be as cost-effective as DFT ones and often slightly more, while showing a satisfactory accuracy when compared with the reference data. Although the focus is on barrier heights, the results may bear broader implications, in that one may see successes and difficulties of DFT when compared with traditional MO theories for the same data.

Integrating sampling techniques and inverse virtual screening: toward the discovery of artificial peptide-based receptors for ligands.

A novel heuristic using an iterative select-and-purge strategy is proposed. It combines statistical techniques for sampling and classification by rigid molecular docking through an inverse virtual screening scheme. This approach aims to the de novo discovery of short peptides that may act as docking receptors for small target molecules when there are no data available about known association complexes between them. The algorithm performs an unbiased stochastic exploration of the sample space, acting as a binary classifier when analyzing the entire peptides population. It uses a novel and effective criterion for weighting the likelihood of a given peptide to form an association complex with a particular ligand molecule based on amino acid sequences. The exploratory analysis relies on chemical information of peptides composition, sequence patterns, and association free energies (docking scores) in order to converge to those peptides forming the association complexes with higher affinities. Statistical estimations support these results providing an association probability by improving predictions accuracy even in cases where only a fraction of all possible combinations are sampled. False positives/false negatives ratio was also improved with this method. A simple rigid-body docking approach together with the proper information about amino acid sequences was used. The methodology was applied in a retrospective docking study to all 8000 possible tripeptide combinations using the 20 natural amino acids, screened against a training set of 77 different ligands with diverse functional groups. Afterward, all tripeptides were screened against a test set of 82 ligands, also containing different functional groups. Results show that our integrated methodology is capable of finding a representative group of the top-scoring tripeptides. The associated probability of identifying the best receptor or a group of the top-ranked receptors is more than double and about 10 times higher, respectively, when compared to classical random sampling methods.

Pyrrolyl-Silicon Compounds as Precursors for Donor-Acceptor Systems Stabilized by Noncovalent Interactions.

Pyrrolyl-silicon compounds were investigated by different theoretical approaches. Model monomers consisted of a pyrrole ring N-substituted with silylmethoxy and silylhydroxy end groups through a propyl chain spacer, designated as PySi and PySiOH. Geometrical, vibrational, and electronic properties, as well as chemical reactivity, are discussed and compared with pyrrole (Py) and N-propylpyrrole (N-PrPy) that were studied in parallel for reference purposes and methods validation. The electronic distribution between PySi and PySiOH differs importantly, the former being an electron donor, as Py and N-PrPy. Conversely, PySiOH presents donor-acceptor character with the LUMO energy level localized on the silanol end group. Global and local reactivity descriptors predict PySiOH more reactive than PySi with two preferential reactive sites: electron-rich Py ring and electron-deficient silanol group. On the basis of experimental studies, oligomers of PySiOH linked α-α' via Py rings (α-α'PynSiOH, n = 2, 3) were considered as model molecules of hydrolyzed PySi. The most stable structures were derived from randomly generated α-α'PynSiOH that were optimized at semiempirical AM1 and refined with M05-2X/6-31G(d,p). Conformational analysis of dimer and trimer structures points to stability enhanced by molecular packing. Nonetheless, NBO and RDG results indicate that oligomer stability is dictated by the cooperative contribution of hydrogen bonding between silanol end groups and dispersive vdW interactions between silanol and the π system of the Py ring. The latter interaction resulting from electron delocalization induced by an electron-deficient silanol group seems to determine the smaller gap energy of T-shaped OH-π arrangements. The theoretical findings support the peculiar chemical behavior revealed by experiment.

Electron density deformations provide new insights into the spectral shift of rhodopsins.

Spectral shifts of rhodopsin, which are related to variations of the electron distribution in 11-cis-retinal, are investigated here using the method of deformed atoms in molecules. We found that systems carrying the M207R and S186W mutations display large perturbations of the π-conjugated system with respect to wild-type rhodopsins. These changes agree with the predicted behavior of the bond length alternation (BLA) and the blue shifts of vertical excitation energies of these systems. The effect of the planarity of the central and Schiff-base regions of retinal chain on the electronic structure of the chromophore is also investigated. By establishing nonlinear polynomial relations between BLA, chain distortions, and vertical excitation energies, we are also able to provide a semiquantitative approach for the understanding of the mechanisms regulating spectral shifts in rhodopsin and its mutants.

Integration of ligand and structure-based virtual screening for identification of leading anabolic steroids.

Parallel ligand- and structure-based virtual screenings of 269 steroids with anabolic activity evaluated in vivo were performed. The quantitative structure-activity relationship (QSAR) model expressed by selected descriptors as the octanol-water partition coefficient, the molar volume and the quantum mechanical calculated charge values on atoms C1, C2, C5, C9, C10, C14 and C17 of the steroid skeleton, expresses structural features of anabolic steroids (AS) contributing to the transport and steroid-receptor interaction. On the other hand, computational simulations of a candidate ligand binding to a receptor study (a "docking" procedure) predict the association of these AS with the human androgen receptor (AR). Fourteen compounds were identified as lead; the most potent was the 7α-methylestr-4-en-3, 17-dione. It was concluded that a good anabolic activity requires hydrogen bonding interactions between both Arg752 and Gln711 residues in the cycles A with O3 atom of the steroid and either Asn705 and Thr877 residues in the cycles D of steroid with O17 atom.