Deep learning on the 2-dimensional Ising model to extract the crossover region with a variational autoencoder.

The 2-dimensional Ising model on a square lattice is investigated with a variational autoencoder in the non-vanishing field case for the purpose of extracting the crossover region between the ferromagnetic and paramagnetic phases. The encoded latent variable space is found to provide suitable metrics for tracking the order and disorder in the Ising configurations that extends to the extraction of a crossover region in a way that is consistent with expectations. The extracted results achieve an exceptional prediction for the critical point as well as agreement with previously published results on the configurational magnetizations of the model. The performance of this method provides encouragement for the use of machine learning to extract meaningful structural information from complex physical systems where little a priori data is available.

Assessing the similarity of ligand binding conformations with the Contact Mode Score.

Structural and computational biologists often need to measure the similarity of ligand binding conformations. The commonly used root-mean-square deviation (RMSD) is not only ligand-size dependent, but also may fail to capture biologically meaningful binding features. To address these issues, we developed the Contact Mode Score (CMS), a new metric to assess the conformational similarity based on intermolecular protein-ligand contacts. The CMS is less dependent on the ligand size and has the ability to include flexible receptors. In order to effectively compare binding poses of non-identical ligands bound to different proteins, we further developed the eXtended Contact Mode Score (XCMS). We believe that CMS and XCMS provide a meaningful assessment of the similarity of ligand binding conformations. CMS and XCMS are freely available at http://brylinski.cct.lsu.edu/content/contact-mode-score and http://geaux-computational-bio.github.io/contact-mode-score/.

GeauxDock: Accelerating Structure-Based Virtual Screening with Heterogeneous Computing.

Computational modeling of drug binding to proteins is an integral component of direct drug design. Particularly, structure-based virtual screening is often used to perform large-scale modeling of putative associations between small organic molecules and their pharmacologically relevant protein targets. Because of a large number of drug candidates to be evaluated, an accurate and fast docking engine is a critical element of virtual screening. Consequently, highly optimized docking codes are of paramount importance for the effectiveness of virtual screening methods. In this communication, we describe the implementation, tuning and performance characteristics of GeauxDock, a recently developed molecular docking program. GeauxDock is built upon the Monte Carlo algorithm and features a novel scoring function combining physics-based energy terms with statistical and knowledge-based potentials. Developed specifically for heterogeneous computing platforms, the current version of GeauxDock can be deployed on modern, multi-core Central Processing Units (CPUs) as well as massively parallel accelerators, Intel Xeon Phi and NVIDIA Graphics Processing Unit (GPU). First, we carried out a thorough performance tuning of the high-level framework and the docking kernel to produce a fast serial code, which was then ported to shared-memory multi-core CPUs yielding a near-ideal scaling. Further, using Xeon Phi gives 1.9× performance improvement over a dual 10-core Xeon CPU, whereas the best GPU accelerator, GeForce GTX 980, achieves a speedup as high as 3.5×. On that account, GeauxDock can take advantage of modern heterogeneous architectures to considerably accelerate structure-based virtual screening applications. GeauxDock is open-sourced and publicly available at www.brylinski.org/geauxdock and https://figshare.com/articles/geauxdock_tar_gz/3205249.

Sign Learning Kink-based (SiLK) Quantum Monte Carlo for molecular systems.

The Sign Learning Kink (SiLK) based Quantum Monte Carlo (QMC) method is used to calculate the ab initio ground state energies for multiple geometries of the H2O, N2, and F2 molecules. The method is based on Feynman's path integral formulation of quantum mechanics and has two stages. The first stage is called the learning stage and reduces the well-known QMC minus sign problem by optimizing the linear combinations of Slater determinants which are used in the second stage, a conventional QMC simulation. The method is tested using different vector spaces and compared to the results of other quantum chemical methods and to exact diagonalization. Our findings demonstrate that the SiLK method is accurate and reduces or eliminates the minus sign problem.

Cooling Atomic Gases With Disorder.

Cold atomic gases have proven capable of emulating a number of fundamental condensed matter phenomena including Bose-Einstein condensation, the Mott transition, Fulde-Ferrell-Larkin-Ovchinnikov pairing, and the quantum Hall effect. Cooling to a low enough temperature to explore magnetism and exotic superconductivity in lattices of fermionic atoms remains a challenge. We propose a method to produce a low temperature gas by preparing it in a disordered potential and following a constant entropy trajectory to deliver the gas into a nondisordered state which exhibits these incompletely understood phases. We show, using quantum Monte Carlo simulations, that we can approach the Néel temperature of the three-dimensional Hubbard model for experimentally achievable parameters. Recent experimental estimates suggest the randomness required lies in a regime where atom transport and equilibration are still robust.

What is the Valence of Mn in Ga(1-x)Mn(x)N?

We investigate the current debate on the Mn valence in Ga(1-x)Mn(x)N, a diluted magnetic semiconductor (DMS) with a potentially high Curie temperature. From a first-principles Wannier-function analysis, we unambiguously find the Mn valence to be close to 2+ (d(5)), but in a mixed spin configuration with average magnetic moments of 4µ(B). By integrating out high-energy degrees of freedom differently, we further derive for the first time from first-principles two low-energy pictures that reflect the intrinsic dual nature of the doped holes in the DMS: (1) an effective d(4) picture ideal for local physics, and (2) an effective d(5) picture suitable for extended properties. In the latter, our results further reveal a few novel physical effects, and pave the way for future realistic studies of magnetism. Our study not only resolves one of the outstanding key controversies of the field, but also exemplifies the general need for multiple effective descriptions to account for the rich low-energy physics in many-body systems in general.

GeauxDock: A novel approach for mixed-resolution ligand docking using a descriptor-based force field.

Molecular docking is an important component of computer-aided drug discovery. In this communication, we describe GeauxDock, a new docking approach that builds on the ideas of ligand homology modeling. GeauxDock features a descriptor-based scoring function integrating evolutionary constraints with physics-based energy terms, a mixed-resolution molecular representation of protein-ligand complexes, and an efficient Monte Carlo sampling protocol. To drive docking simulations toward experimental conformations, the scoring function was carefully optimized to produce a correlation between the total pseudoenergy and the native-likeness of binding poses. Indeed, benchmarking calculations demonstrate that GeauxDock has a strong capacity to identify near-native conformations across docking trajectories with the area under receiver operating characteristics of 0.85. By excluding closely related templates, we show that GeauxDock maintains its accuracy at lower levels of homology through the increased contribution from physics-based energy terms compensating for weak evolutionary constraints. GeauxDock is available at http://www.institute.loni.org/lasigma/package/dock/.

Accelerating the Pace of Protein Functional Annotation With Intel Xeon Phi Coprocessors.

Intel Xeon Phi is a new addition to the family of powerful parallel accelerators. The range of its potential applications in computationally driven research is broad; however, at present, the repository of scientific codes is still relatively limited. In this study, we describe the development and benchmarking of a parallel version of eFindSite, a structural bioinformatics algorithm for the prediction of ligand-binding sites in proteins. Implemented for the Intel Xeon Phi platform, the parallelization of the structure alignment portion of eFindSite using pragma-based OpenMP brings about the desired performance improvements, which scale well with the number of computing cores. Compared to a serial version, the parallel code runs 11.8 and 10.1 times faster on the CPU and the coprocessor, respectively; when both resources are utilized simultaneously, the speedup is 17.6. For example, ligand-binding predictions for 501 benchmarking proteins are completed in 2.1 hours on a single Stampede node equipped with the Intel Xeon Phi card compared to 3.1 hours without the accelerator and 36.8 hours required by a serial version. In addition to the satisfactory parallel performance, porting existing scientific codes to the Intel Xeon Phi architecture is relatively straightforward with a short development time due to the support of common parallel programming models by the coprocessor. The parallel version of eFindSite is freely available to the academic community at www.brylinski.org/efindsite.

Equation of motion coupled cluster methods for electron attachment and ionization potential in fullerenes C60 and C70.

In both molecular and periodic solid-state systems there is a need for the accurate determination of the ionization potential and the electron affinity for systems ranging from light harvesting polymers and photocatalytic compounds to semiconductors. The development of a Green's function approach based on the coupled cluster (CC) formalism would be a valuable tool for addressing many properties involving many-body interactions along with their associated correlation functions. As a first step in this direction, we have developed an accurate and parallel efficient approach based on the equation of motion-CC technique. To demonstrate the high degree of accuracy and numerical efficiency of our approach we calculate the ionization potential and electron affinity for C60 and C70. Accurate predictions for these molecules are well beyond traditional molecular scale studies. We compare our results with experiments and both quantum Monte Carlo and GW calculations.

Effective cluster typical medium theory for the diagonal Anderson disorder model in one- and two-dimensions.

We develop a cluster typical medium theory to study localization in disordered electronic systems. Our formalism is able to incorporate non-local correlations beyond the local typical medium theory in a systematic way. The cluster typical medium theory utilizes the momentum-resolved typical density of states and hybridization function to characterize the localization transition. We apply the formalism to the Anderson model of localization in one- and two-dimensions. In one-dimension, we find that the critical disorder strength scales inversely with the linear cluster size with a power law, Wc ∼ (1/Lc)(1/ν), whereas in two-dimensions, the critical disorder strength decreases logarithmically with the linear cluster size. Our results are consistent with previous numerical work and are in agreement with the one-parameter scaling theory.

First-principles Wannier function analysis of the electronic structure of PdTe: weaker magnetism and superconductivity.

We report a first-principles Wannier function study of the electronic structure of PdTe. Its electronic structure is found to be a broad three-dimensional Fermi surface with highly reduced correlation effects. In addition, the higher filling of the Pd d-shell, its stronger covalency resulting from the closer energy of the Pd d and Te p shells, and the larger crystal field effects of the Pd ion due to its near octahedral coordination, all serve to weaken significantly electronic correlations in the particle-hole (spin, charge, and orbital) channel. In comparison to the Fe chalcogenides, e.g. FeSe, we highlight the essential features (quasi-two-dimensionality, proximity to half-filling, weaker covalency, and higher orbital degeneracy) of Fe-based high-temperature superconductors.

Unconventional superconductivity from local spin fluctuations in the Kondo lattice.

The explanation of heavy-fermion superconductivity is a long-standing challenge to theory. It is commonly thought to be connected to nonlocal fluctuations of either spin or charge degrees of freedom and therefore of unconventional type. Here we present results for the Kondo-lattice model, a paradigmatic model to describe heavy-fermion compounds, obtained from dynamical mean-field theory which captures local correlation effects only. Unexpectedly, we find robust s-wave superconductivity in the heavy-fermion state. We argue that this novel type of pairing is tightly connected to the formation of heavy quasiparticle bands and the presence of strong local spin fluctuations.

Analytic continuation of quantum Monte Carlo data by stochastic analytical inference.

We present an algorithm for the analytic continuation of imaginary-time quantum Monte Carlo data which is strictly based on principles of Bayesian statistical inference. Within this framework we are able to obtain an explicit expression for the calculation of a weighted average over possible energy spectra, which can be evaluated by standard Monte Carlo simulations, yielding as by-product also the distribution function as function of the regularization parameter. Our algorithm thus avoids the usual ad hoc assumptions introduced in similar algorithms to fix the regularization parameter. We apply the algorithm to imaginary-time quantum Monte Carlo data and compare the resulting energy spectra with those from a standard maximum-entropy calculation.

Continuous-time quantum Monte Carlo and maximum entropy approach to an imaginary-time formulation of strongly correlated steady-state transport.

Recently, Han and Heary [Phys. Rev. Lett. 99, 236808 (2007)] proposed an approach to steady-state quantum transport through mesoscopic structures, which maps the nonequilibrium problem onto a family of auxiliary quantum impurity systems subject to imaginary voltages. We employ continuous-time quantum Monte-Carlo solvers to calculate accurate imaginary time data for the auxiliary models. The spectral function is obtained from a maximum entropy analytical continuation in both Matsubara frequency and complexified voltage. To enable the analytical continuation we construct a kernel which is compatible with the analytical structure of the theory. While it remains a formidable task to extract reliable spectral functions from this unbiased procedure, particularly for large voltages, our results indicate that the method in principle yields results in agreement with those obtained by other methods.

Nonlocal effects on magnetism in the diluted magnetic semiconductor Ga1-xMnxAs.

The magnetic properties of the diluted magnetic semiconductor Ga1-xMnxAs are studied within the dynamical cluster approximation. We use the k x p Hamiltonian to describe the electronic structure of GaAs with spin-orbit coupling and strain effects. We show that nonlocal effects are essential for explaining the experimentally observed transition temperature and saturation magnetization. We also demonstrate that the cluster anisotropy is very strong and induces rotational frustration and a cube-edge direction magnetic anisotropy at low temperature. With this, we explain the temperature-driven spin reorientation in this system.

Transition temperature of a magnetic semiconductor with angular momentum j.

We employ dynamical mean-field theory to identify the materials properties that optimize T(c) for a generalized double-exchange model. We reach the surprising conclusion that T(c) achieves a maximum when the band angular momentum j equals 3/2 and when the masses in the m(j) = +/- 1/2 and +/-3/2 and subbands are equal. However, we also find that T(c) is significantly reduced as the ratio of the masses decreases from one. Consequently, the search for dilute-magnetic semiconductor materials with high T(c) should proceed on two fronts. In semiconductors with p bands, such as the currently studied Mn-doped Ge and GaAs semiconductors, T(c) may be optimized by tuning the band masses through strain engineering or artificial nanostructures. On the other hand, semiconductors with s or d bands with nearly equal effective masses might prove to have higher T(c)'s than p-band materials with disparate effective masses.

Enhanced recognition of protein transmembrane domains with prediction-based structural profiles.

MOTIVATION: Membrane domain prediction has recently been re-evaluated by several groups, suggesting that the accuracy of existing methods is still rather limited. In this work, we revisit this problem and propose novel methods for prediction of alpha-helical as well as beta-sheet transmembrane (TM) domains. The new approach is based on a compact representation of an amino acid residue and its environment, which consists of predicted solvent accessibility and secondary structure of each amino acid. A recently introduced method for solvent accessibility prediction trained on a set of soluble proteins is used here to indicate segments of residues that are predicted not to be accessible to water and, therefore, may be 'buried' in the membrane. While evolutionary profiles in the form of a multiple alignment are used to derive these simple 'structural profiles', they are not used explicitly for the membrane domain prediction and the overall number of parameters in the model is significantly reduced. This offers the possibility of a more reliable estimation of the free parameters in the model with a limited number of experimentally resolved membrane protein structures. RESULTS: Using cross-validated training on available sets of structurally resolved and non-redundant alpha and beta membrane proteins, we demonstrate that membrane domain prediction methods based on such a compact representation outperform approaches that utilize explicitly evolutionary profiles and multiple alignments. Moreover, using an external evaluation by the TMH Benchmark server we show that our final prediction protocol for the TM helix prediction is competitive with the state-of-the-art methods, achieving per-residue accuracy of approximately 89% and per-segment accuracy of approximately 80% on the set of high resolution structures used by the TMH Benchmark server. At the same time the observed rates of confusion with signal peptides and globular proteins are the lowest among the tested methods. The new method is available online at http://minnou.cchmc.org.

Weekly ECT in Geriatric Depression.

Fifteen elderly depressed psychiatric inpatients were randomly assigned to receive either standard three-times-weekly electroconvulsive therapy (ECT) or once-weekly ECT. Outcome measures included cognitive assessment and antidepressant response. Although both groups improved with treatment, the three-times-weekly group improved substantially more quickly. There was no difference in cognitive effect between the two groups. We conclude that the traditional three-times-weekly schedule of ECT may optimally balance speed of antidepressant response and cognitive impairment.