Artificial intelligence for the prevention and prediction of colorectal neoplasms.

BACKGROUND: Colonoscopy is a useful as a cancer screening test. However, in countries with limited medical resources, there are restrictions on the widespread use of endoscopy. Non-invasive screening methods to determine whether a patient requires a colonoscopy are thus desired. Here, we investigated whether artificial intelligence (AI) can predict colorectal neoplasia. METHODS: We used data from physical exams and blood analyses to determine the incidence of colorectal polyp. However, these features exhibit highly overlapping classes. The use of a kernel density estimator (KDE)-based transformation improved the separability of both classes. RESULTS: Along with an adequate polyp size threshold, the optimal machine learning (ML) models' performance provided 0.37 and 0.39 Matthews correlation coefficient (MCC) for the datasets of men and women, respectively. The models exhibit a higher discrimination than fecal occult blood test with 0.047 and 0.074 MCC for men and women, respectively. CONCLUSION: The ML model can be chosen according to the desired polyp size discrimination threshold, may suggest further colorectal screening, and possible adenoma size. The KDE feature transformation could serve to score each biomarker and background factors (health lifestyles) to suggest measures to be taken against colorectal adenoma growth. All the information that the AI model provides can lower the workload for healthcare providers and be implemented in health care systems with scarce resources. Furthermore, risk stratification may help us to optimize the efficiency of resources for screening colonoscopy.

De novo generation of optically active small organic molecules using Monte Carlo tree search combined with recurrent neural network.

Optically active small organic molecules are computationally designed using the ChemTS python library developed by Tsuda and collaborators, which utilizes a combined Monte Carlo tree search (MCTS) and recurrent neural network model. Geometry optimization and excited-state calculations are performed for each generated molecule, following which the excitation energy and dissymmetry factors are computed to evaluate the score function in the MCTS process. Using this procedure, molecules not contained in existing databases are generated. Molecules having either high dissymmetry factors or high transition dipole strengths can be generated depending on the choice of the score function. In a single trajectory with 100,000 trials, mutually similar high-scoring molecules are generated frequently after the initial 15,000-20,000 trials. This indicates that it is better to sample high-scoring molecules from several trajectories having a modest number of trials each than from a single trajectory having a large number of trials.

Anisotropic Poisson Effect and Deformation-Induced Fluorescence Change of Elastic 9,10-Dibromoanthracene Single Crystals.

Elastic organic crystals have attracted considerable attention as next-generation flexible smart materials. However, the detailed information on both molecular packing change and macroscopic mechanical crystal deformations upon applied stress is still insufficient. Herein, we report that fluorescent single crystals of 9,10-dibromoanthracene are elastically bendable and stretchable, which allows a detailed investigation of the deformation behavior. We clearly observed a Poisson effect for the crystal, where the short axes (b and c-axes) of the crystal are contracted upon elongation along the long axis (a-axis). Moreover, we found that the Poisson's ratios along the b-axis and c-axis are largely different. Theoretical molecular simulation suggests that the tilting motion of the anthracene may be responsible for the large deformation along the c-axis. Spatially resolved photoluminescence (PL) measurement of the bent elastic crystals reveals that the PL spectra at the outer (elongated), central (neutral), and inner (contracted) sides are different from each other.

New Insights in the Natural Organic Matter Fouling Mechanism of Polyamide and Nanocomposite Multiwalled Carbon Nanotubes-Polyamide Membranes.

Polyamide (PA) membranes comprise most of the reverse osmosis membranes currently used for desalination and water purification. However, their fouling mechanisms with natural organic matter (NOM) is still not completely understood. In this work, we studied three different types of PA membranes: a laboratory made PA, a commercial PA, and a multiwalled carbon nanotube (CNT-PA nanocomposite membrane during cross-flow measurements by NaCl solutions including NOM, humic acid (HA), or alginate, respectively). Molecular dynamic simulations were also used to understand the fouling process of NOM down to its molecular scale. Low molecular weight humic acid binds to the surface cavities on the PA structures that leads to irreversible adsorption induced by the high surface roughness. In addition, the larger alginate molecules show a different mechanism, due to their larger size and their ability to change shape from the globule type to the uncoiled state. Specifically, alginate molecules either bind through Ca2+ bridges or they uncoil and spread on the surface. This work shows that carbon nanotubes can help to decrease roughness and polymer mobility on the surfaces of the membranes at the molecular scale, which represents a novel method to design antifouling membranes.

Inversion of Optical Activity in the Synthesis of Mercury Sulfide Nanoparticles: Role of Ligand Coordination.

Optical activity in inorganic colloidal materials was controlled through interactions of chiral molecules with the nanoparticle (NP) surface. An inversion of optical activity in the synthesis of mercury sulfide (HgS) NPs was demonstrated with an intrinsically chiral crystalline system in the presence of an identical chiral capping ligand. A continuous decrease in the positive first Cotton effect and an eventual reversal of CD profile were observed upon heating the aqueous solution of HgS NPs capped with N-acetyl-l-cysteine (Ac-l-Cys) at 80 °C. Ac-l-Cys afforded two bidentate coordination configurations with an almost mirror image of each other using the thiolate and either of carboxylate or acetyl-carbonyl groups on the HgS core. Experiment and calculation suggest that a shift in the distribution of the NP formation with energy in response to the combinations of ligand coordination structure and chiral crystalline surface is responsible for the inversion of optical activity.

MPI/OpenMP hybrid parallel algorithm for resolution of identity second-order Møller-Plesset perturbation calculation of analytical energy gradient for massively parallel multicore supercomputers.

A massively parallel algorithm of the analytical energy gradient calculations based the resolution of identity Møller-Plesset perturbation (RI-MP2) method from the restricted Hartree-Fock reference is presented for geometry optimization calculations and one-electron property calculations of large molecules. This algorithm is designed for massively parallel computation on multicore supercomputers applying the Message Passing Interface (MPI) and Open Multi-Processing (OpenMP) hybrid parallel programming model. In this algorithm, the two-dimensional hierarchical MP2 parallelization scheme is applied using a huge number of MPI processes (more than 1000 MPI processes) for acceleration of the computationally demanding O(N5 ) step such as calculations of occupied-occupied and virtual-virtual blocks of MP2 one-particle density matrix and MP2 two-particle density matrices. The new parallel algorithm performance is assessed using test calculations of several large molecules such as buckycatcher C60 @C60 H28 (144 atoms, 1820 atomic orbitals (AOs) for def2-SVP basis set, and 3888 AOs for def2-TZVP), nanographene dimer (C96 H24 )2 (240 atoms, 2928 AOs for def2-SVP, and 6432 AOs for cc-pVTZ), and trp-cage protein 1L2Y (304 atoms and 2906 AOs for def2-SVP) using up to 32,768 nodes and 262,144 central processing unit (CPU) cores of the K computer. The results of geometry optimization calculations of trp-cage protein 1L2Y at the RI-MP2/def2-SVP level using the 3072 nodes and 24,576 cores of the K computer are presented and discussed to assess the efficiency of the proposed algorithm. © 2017 Wiley Periodicals, Inc.

From C60 to Infinity: Large-Scale Quantum Chemistry Calculations of the Heats of Formation of Higher Fullerenes.

We have carried out large-scale computational quantum chemistry calculations on the K computer to obtain heats of formation for C60 and some higher fullerenes with the DSD-PBE-PBE/cc-pVQZ double-hybrid density functional theory method. Our best estimated values are 2520.0 ± 20.7 (C60), 2683.4 ± 17.7 (C70), 2862.0 ± 18.5 (C76), 2878.8 ± 13.3 (C78), 2946.4 ± 14.5 (C84), 3067.3 ± 15.4 (C90), 3156.6 ± 16.2 (C96), 3967.7 ± 33.4 (C180), 4364 (C240) and 5415 (C320) kJ mol(-1). In our assessment, we also find that the B3-PW91-D3BJ and BMK-D3(BJ) functionals perform reasonably well. Using the convergence behavior for the calculated per-atom heats of formation, we obtained the formula ΔfH per carbon = 722n(-0.72) + 5.2 kJ mol(-1) (n = the number of carbon atoms), which enables an estimation of ΔfH for higher fullerenes more generally. A slow convergence to the graphene limit is observed, which we attribute to the relatively small proportion of fullerene carbons that are in "low-strain" regions. We further propose that it would take tens, if not hundreds, of thousands of carbons for a fullerene to roughly approach the limit. Such a distinction may be a contributing factor to the discrete properties between the two types of nanomaterials. During the course of our study, we also observe a fairly reliable means for the theoretical calculation of heats of formation for medium-sized fullerenes. This involves the use of isodesmic-type reactions with fullerenes of similar sizes to provide a good balance of the chemistry and to minimize the use of accompanying species.

A Series of Layered Assemblies of Hydrogen-Bonded, Hexagonal Networks of C3-Symmetric π-Conjugated Molecules: A Potential Motif of Porous Organic Materials.

Hydrogen-bonded porous organic crystals are promising candidates for functional organic materials due to their easy construction and flexibility arising from reversible bond formation-dissociation. However, it still remains challenging to form porous materials with void spaces that are well-controlled in size, shape, and multiplicity because even well-designed porous frameworks often fail to generate pores within the crystal due to unexpected disruption of hydrogen bonding networks or interpenetration of the frameworks. Herein, we demonstrate that a series of C3-symmetric π-conjugated planar molecules (Tp, T12, T18, and Ex) with three 4,4'-dicarboxy-o-terphenyl moieties in their periphery can form robust hydrogen-bonded hexagonal networks (H-HexNets) with dual or triple pores and that the H-HexNets stack without interpenetration to yield a layered assembly of H-HexNet (LA-H-HexNet) with accessible volumes up to 59%. Specifically, LA-H-HexNets of Tp and T12 exhibit high crystallinity and permanent porosity after desolvation (activation): SABET = 788 and 557 m(2) g(-1), respectively, based on CO2 sorption at 195 K. We believe that the present design principle can be applied to construct a wide range of two-dimensional noncovalent organic frameworks (2D-nCOFs) and create a pathway to the development of a new class of highly porous functional materials.

Massively parallel algorithm and implementation of RI-MP2 energy calculation for peta-scale many-core supercomputers.

A new parallel algorithm and its implementation for the RI-MP2 energy calculation utilizing peta-flop-class many-core supercomputers are presented. Some improvements from the previous algorithm (J. Chem. Theory Comput. 2013, 9, 5373) have been performed: (1) a dual-level hierarchical parallelization scheme that enables the use of more than 10,000 Message Passing Interface (MPI) processes and (2) a new data communication scheme that reduces network communication overhead. A multi-node and multi-GPU implementation of the present algorithm is presented for calculations on a central processing unit (CPU)/graphics processing unit (GPU) hybrid supercomputer. Benchmark results of the new algorithm and its implementation using the K computer (CPU clustering system) and TSUBAME 2.5 (CPU/GPU hybrid system) demonstrate high efficiency. The peak performance of 3.1 PFLOPS is attained using 80,199 nodes of the K computer. The peak performance of the multi-node and multi-GPU implementation is 514 TFLOPS using 1349 nodes and 4047 GPUs of TSUBAME 2.5. © 2016 Wiley Periodicals, Inc.

Optimization of RI-MP2 auxiliary basis functions for 6-31G** and 6-311G** basis sets for first-, second-, and third-row elements.

Auxiliary basis functions for second-order Møller-Plesset perturbation theory with resolution-of-identity approximation (RI-MP2) are developed for first-, second-, and third-row elements, which are suitable for Pople-type 6-31G** and 6-311G** basis sets. Atomic-centered Gaussian functions up to the g-type function are used for auxiliary basis functions to obtain higher accuracy for molecules with the accurate description of bonding properties. The performance of the developed auxiliary basis functions were tested and evaluated for 114 small and 23 large molecules. The developed auxiliary basis functions show much smaller energy differences between MP2 and RI-MP2 than other auxiliary basis functions used for 6-31G** and 6-311G** basis sets with similar computational costs.

Two-level hierarchical parallelization of second-order Møller-Plesset perturbation calculations in divide-and-conquer method.

A two-level hierarchical parallelization scheme including the second-order Møller-Plesset perturbation (MP2) theory in the divide-and-conquer method is presented. The scheme is a combination of coarse-grain parallelization assigning each subsystem to a group of processors, with fine-grain parallelization, where the computational tasks for evaluating MP2 correlation energy of the assigned subsystem are distributed among processors in the group. Test calculations demonstrate that the present scheme shows high parallel efficiency and makes MP2 calculations practical for very large molecules.

Efficient Search for Energetically Favorable Molecular Conformations against Metastable States via Gray-Box Optimization.

In order to accurately understand and estimate molecular properties, finding energetically favorable molecular conformations is the most fundamental task for atomistic computational research on molecules and materials. Geometry optimization based on quantum chemical calculations has enabled the conformation prediction of arbitrary molecules, including de novo ones. However, it is computationally expensive to perform geometry optimizations for enormous conformers. In this study, we introduce the gray-box optimization (GBO) framework, which enables optimal control over the entire geometry optimization process, among multiple conformers. Algorithms designed for GBO roughly estimate energetically preferable conformers during their geometry optimization iterations. They then preferentially compute promising conformers. To evaluate the performance of the GBO framework, we applied it to a test set consisting of seven dipeptides and mycophenolic acid to determine their stable conformations at the density functional theory level. We thus preferentially obtained energetically favorable conformations. Furthermore, the computational costs required to find the most stable conformation were significantly reduced (approximately 1% on average, compared to the naive approach for the dipeptides).

Molecular tailoring approach in conjunction with MP2 and Ri-MP2 codes: A comparison with fragment molecular orbital method.

Many Divide-and-Conquer based approaches are being developed to overcome the high scaling problem of the ab initio methods. In this work, one such method, Molecular Tailoring Approach (MTA) has been interfaced with recently developed efficient Møller-Plesset second order perturbation theory (MP2) codes viz. IMS-MP2 and RI-MP2 to reap the advantage of both. An external driver script is developed for implementing MTA at the front-end and the MP2 codes at the back-end. The present version of the driver script is written only for a single point energy evaluation of a molecular system at a fixed geometry. The performance of these newly developed MTA-IMS-MP2 and MTA-RI-MP2 codes is extensively benchmarked for a variety of molecular systems vis-à-vis the corresponding actual runs. In addition to this, the performance of these programs is also critically compared with Fragment Molecular Orbital (FMO), another popular fragment-based method. It is observed that FMO2/2 is superior to FMO3/2 and MTA with respect to time advantage; however, the errors of FMO2 are much beyond chemical accuracy. However, FMO3/2 is a highly accurate method for biological systems but is unsuccessful in case of water clusters. MTA produces estimates with errors within 1 kcal/mol uniformly for all systems with reasonable time advantage. Analysis carried out employing various basis sets shows that FMO gives its optimum performance only for basis sets, which does not include diffuse functions. On the contrary, MTA performance is found to be similar for any basis set used.

Application of second-order Møller-Plesset perturbation theory with resolution-of-identity approximation to periodic systems.

Efficient periodic boundary condition (PBC) calculations by the second-order Møller-Plesset perturbation (MP2) method based on crystal orbital formalism are developed by introducing the resolution-of-identity (RI) approximation of four-center two-electron repulsion integrals (ERIs). The formulation and implementation of the PBC RI-MP2 method are presented. In this method, the mixed auxiliary basis functions of the combination of Poisson and Gaussian type functions are used to circumvent the slow convergence of the lattice sum of the long-range ERIs. Test calculations of one-dimensional periodic trans-polyacetylene show that the PBC RI-MP2 method greatly reduces the computational times as well as memory and disk sizes, without the loss of accuracy, compared to the conventional PBC MP2 method.

A Simple Model for Relative Energies of All Fullerenes Reveals the Interplay between Intrinsic Resonance and Structural Deformation Effects in Medium-Sized Fullerenes.

Fullerenes are sheets of sp2 carbon atoms wrapped around to form spheres. With this simple consideration, we have in the present study devised and (with over 3600 DFT data points) successfully validated a simple model, termed R+D, for estimating the relative energies of fullerenes. This model contains a resonance component to account for the intrinsic differences between the π-energies of different fullerenes, and a deformation component for treating the distortions from planarity. Notably, we find that both terms (and they alone) are required to obtain good relative energies, which lends support to the formulation of the R+D model. An interesting finding is that for some medium-sized IPR fullerenes, their isomers show similar variations in the two components. We deduce that these fullerenes may represent a good opportunity for tuning molecular properties for practical applications. We hope that the promising results of the present study will encourage further investigations into fullerenes from a fundamental perspective.

From Linear to Foldamer and Assembly: Hierarchical Transformation of a Coplanar Conjugated Polymer into a Microsphere.

Despite the coplanar structure, a conjugated alternating copolymer forms amorphous, well-defined microspheres without π-stacked crystalline domains. Here, we gain insights into the mechanism of how the coplanar conjugated polymer forms amorphous microspheres by means of spectroscopic studies on the assembly/disassembly processes. The difference of the spectral profiles of photoabsorption and photoluminescence with varying solvent/nonsolvent composition clarifies that stepwise assembly takes place through the microsphere formation; [1] intrapolymer linear-to-folding transformation upon diffusion of polar nonsolvent and [2] interpolymer assembly of the foldamers upon further addition of the nonsolvent to form microspheres. As shown in various biopolymers such as proteins and DNA, such stepwise folding and assembly behaviors of conjugated polymers from primary to secondary and tertiary structure open a new way to create transformable functional materials.

MPI/OpenMP Hybrid Parallel Algorithm of Resolution of Identity Second-Order Møller-Plesset Perturbation Calculation for Massively Parallel Multicore Supercomputers.

A new algorithm for massively parallel calculations of electron correlation energy of large molecules based on the resolution of identity second-order Møller-Plesset perturbation (RI-MP2) technique is developed and implemented into the quantum chemistry software NTChem. In this algorithm, a Message Passing Interface (MPI) and Open Multi-Processing (OpenMP) hybrid parallel programming model is applied to attain efficient parallel performance on massively parallel supercomputers. An in-core storage scheme of intermediate data of three-center electron repulsion integrals utilizing the distributed memory is developed to eliminate input/output (I/O) overhead. The parallel performance of the algorithm is tested on massively parallel supercomputers such as the K computer (using up to 45 992 central processing unit (CPU) cores) and a commodity Intel Xeon cluster (using up to 8192 CPU cores). The parallel RI-MP2/cc-pVTZ calculation of two-layer nanographene sheets (C150H30)2 (number of atomic orbitals is 9640) is performed using 8991 node and 71 288 CPU cores of the K computer.

Energy density analysis of cluster size dependence of surface-molecule interactions: H2, C2H2, C2H4, and CO adsorption onto Si(100)-(2x1) surface.

Adsorption of H2, C2H2, C2H4, and CO onto a Si(100)-(2x1) surface has been treated theoretically using Si(12n - 3)H(8n + 4) (n = 1-4) clusters. The energy density analysis (EDA) proposed by Nakai has been adopted to examine surface-molecule interactions for different cluster sizes. EDA results for the largest model cluster Si45H36 have shown that the adsorption-induced energy density variation in Si atoms decays with distance from the adsorption site. Analysis of this decay, which can be carried out using the EDA technique, is important because it enables verification of the reliability of the model cluster used. In the cases of H2, C2H2, C2H4, and CO adsorption onto the Si(100)-(2x1) surface, it is found that at least a Si21H20 cluster is necessary to treat the surface-molecule interaction with chemical accuracy.