Transferable MP2-Based Machine Learning for Accurate Coupled-Cluster Energies.

Machine learning methods have enabled the low-cost evaluation of molecular properties such as energy at an unprecedented scale. While many of such applications have focused on molecular input based on geometry, few studies consider representations based on the underlying electronic structure. Directing the attention to the electronic structure offers a unique challenge that allows for a more detailed representation of the underlying physics and how they affect molecular properties. The target of this work is to efficiently encode a lower-cost correlated wave function derived from MP2 to predict a higher-cost coupled-cluster singles-and-doubles (CCSD) wave function based on correlation-pair energies and the contributing electron promotions (excitations) and integrals. The new molecular representation explores the short-range behavior of electron correlation and utilizes distinct models that differentiate between two-electron promotions from the same molecular orbital or from two different orbitals. We present a re-engineered set of input features that provide an intuitive description of the orbital properties involved in electron correlation. The overall models are found to be highly transferable and size extensive, necessitating very few training instances to approach the chemical accuracy of a broad spectrum of organic molecules. The efficiency and transferability of the novel representation are demonstrated on a series of linear hydrocarbons, the potential energy surface of the water dimer, and on the GDB-9 database. For the GDB-9 database, we found that data from only 140 randomly selected molecules are adequate to achieve chemical accuracy for more than 133 000 organic molecules.

Author Correction: Representation of molecular structures with persistent homology for machine learning applications in chemistry.

An amendment to this paper has been published and can be accessed via a link at the top of the paper.

Representation of molecular structures with persistent homology for machine learning applications in chemistry.

Machine learning and high-throughput computational screening have been valuable tools in accelerated first-principles screening for the discovery of the next generation of functionalized molecules and materials. The application of machine learning for chemical applications requires the conversion of molecular structures to a machine-readable format known as a molecular representation. The choice of such representations impacts the performance and outcomes of chemical machine learning methods. Herein, we present a new concise molecular representation derived from persistent homology, an applied branch of mathematics. We have demonstrated its applicability in a high-throughput computational screening of a large molecular database (GDB-9) with more than 133,000 organic molecules. Our target is to identify novel molecules that selectively interact with CO2. The methodology and performance of the novel molecular fingerprinting method is presented and the new chemically-driven persistence image representation is used to screen the GDB-9 database to suggest molecules and/or functional groups with enhanced properties.

CO2 Capture on Functionalized Calixarenes: A Computational Study.

High carbon emissions have shown a strong correlation with rising global temperatures as the world's climate undergoes a dramatic shift. Work to mitigate the potential damage using materials such as metal-organic frameworks (MOFs), covalent organic frameworks (COFs), and polymer membranes (PMs) has proven successful in small-scale approaches; however, research is still being performed to enhance the capabilities of these materials for use at an industrial scale. One strategy for increasing performance is to embed these materials with CO2-philic molecules, which enhance selective binding over other gases. Calixarenes are promising candidates due to their large chalice shape, which allows for the possibility to bind multiple CO2 molecules per site. In this study, a dataset including 40 functionalized calixarene structures and one unfunctionalized (bare) calixarene was constructed with an automated, high-throughput structure generation through directed modifications to a molecular scaffold. A conformational search based on molecular mechanics allowed the faster determination of optimal binding energies for a vast array of chemical functional groups with less computational effort. Density functional theory and symmetry-adapted perturbation theory calculations were performed for the exploration of their interactions with CO2. Our work has identified new organic cages with increased CO2-philicity. In four cases, CO2 binding is stronger than 9.0 kcal/mol and very close to the targets set by previous studies. The nature of the noncovalent interactions for these cases is analyzed and discussed. Conclusions from this study can aid synthetic efforts for the next generation of functional materials.

Data-Driven Acceleration of the Coupled-Cluster Singles and Doubles Iterative Solver.

Solving the coupled-cluster (CC) equations is a cost-prohibitive process that exhibits poor scaling with system size. These equations are solved by determining the set of amplitudes (t) that minimize the system energy with respect to the coupled-cluster equations at the selected level of truncation. Here, a novel approach to predict the converged coupled-cluster singles and doubles (CCSD) amplitudes, thus the coupled-cluster wave function, is explored by using machine learning and electronic structure properties inherent to the MP2 level. Features are collected from quantum chemical data, such as orbital energies, one-electron Hamiltonian, Coulomb, and exchange terms. The data-driven CCSD (DDCCSD) is not an alchemical method because the actual iterative coupled-cluster equations are solved. However, accurate energetics can also be obtained by bypassing solving the CC equations entirely. Our preliminary data show that it is possible to achieve remarkable speedups in solving the CCSD equations, especially when the correct physics are encoded and used for training of machine learning models.

Unprecedented Five-Coordinate Iron(IV) Imides Generate Divergent Spin States Based on the Imide R-Groups.

Three five-coordinate iron(IV) imide complexes have been synthesized and characterized. These novel structures have disparate spin states on the iron as a function of the R-group attached to the imide, with alkyl groups leading to low-spin diamagnetic (S=0) complexes and an aryl group leading to an intermediate-spin (S=1) complex. The different spin states lead to significant differences in the bonding about the iron center as well as the spectroscopic properties of these complexes. Mössbauer spectroscopy confirmed that all three imide complexes are in the iron(IV) oxidation state. The combination of diamagnetism and 15 N labeling allowed for the first 15 N NMR resonance recorded on an iron imide. Multi-reference calculations corroborate the experimental structural findings and suggest how the bonding is distinctly different on the imide ligand between the two spin states.

Understanding the Nature of Weak Interactions between Functionalized Boranes and N2/O2, Promising Functional Groups for Gas Separations.

The separation of nitrogen and oxygen gases is considered as a very challenging process, since both O2 and N2 are nonpolar molecules with similar kinetic diameters. Electronic structure theory can provide a fundamental understanding of effects that can lead to selective binding of nitrogen or oxygen gas for the development of novel separation processes. Boranes can bind dinitrogen through a dative bond, where the boron acts as a σ acceptor and back-donates through π orbitals. To better understand these interactions, we have performed highly accurate CCSD(F12)(T) and CCSDT(Q) computations for the BH3-N2 and BH3-O2 complexes. The coupled-cluster binding energies were used as reference for benchmarking different density functionals, and larger functionalized boranes were examined at the M05/def2-TZVPPD level. Symmetry adapted perturbation theory (SAPT) calculations were performed for the elucidation of the nature of the interaction between nitrogen and substituted boranes and how direct or distal functionalizations affect the strength of the weak dative bonds. By use of these methods, several boranes were found to bind N2 over O2. These molecular species are promising functional groups for incorporation into the next generation of advanced materials for efficient N2/O2 separations.

Computational Approach to Molecular Catalysis by 3d Transition Metals: Challenges and Opportunities.

Computational chemistry provides a versatile toolbox for studying mechanistic details of catalytic reactions and holds promise to deliver practical strategies to enable the rational in silico catalyst design. The versatile reactivity and nontrivial electronic structure effects, common for systems based on 3d transition metals, introduce additional complexity that may represent a particular challenge to the standard computational strategies. In this review, we discuss the challenges and capabilities of modern electronic structure methods for studying the reaction mechanisms promoted by 3d transition metal molecular catalysts. Particular focus will be placed on the ways of addressing the multiconfigurational problem in electronic structure calculations and the role of expert bias in the practical utilization of the available methods. The development of density functionals designed to address transition metals is also discussed. Special emphasis is placed on the methods that account for solvation effects and the multicomponent nature of practical catalytic systems. This is followed by an overview of recent computational studies addressing the mechanistic complexity of catalytic processes by molecular catalysts based on 3d metals. Cases that involve noninnocent ligands, multicomponent reaction systems, metal-ligand and metal-metal cooperativity, as well as modeling complex catalytic systems such as metal-organic frameworks are presented. Conventionally, computational studies on catalytic mechanisms are heavily dependent on the chemical intuition and expert input of the researcher. Recent developments in advanced automated methods for reaction path analysis hold promise for eliminating such human-bias from computational catalysis studies. A brief overview of these approaches is presented in the final section of the review. The paper is closed with general concluding remarks.

Diagnostic yield and accuracy of coronary CT angiography after abnormal nuclear myocardial perfusion imaging.

We aimed to determine the diagnostic yield and accuracy of coronary CT angiography (CCTA) in patients referred for invasive coronary angiography (ICA) based on clinical concern for coronary artery disease (CAD) and an abnormal nuclear stress myocardial perfusion imaging (MPI) study. We enrolled 100 patients (84 male, mean age 59.6 ± 8.9 years) with an abnormal MPI study and subsequent referral for ICA. Each patient underwent CCTA prior to ICA. We analyzed the prevalence of potentially obstructive CAD (≥50% stenosis) on CCTA and calculated the diagnostic accuracy of ≥50% stenosis on CCTA for the detection of clinically significant CAD on ICA (defined as any ≥70% stenosis or ≥50% left main stenosis). On CCTA, 54 patients had at least one ≥50% stenosis. With ICA, 45 patients demonstrated clinically significant CAD. A positive CCTA had 100% sensitivity and 84% specificity with a 100% negative predictive value and 83% positive predictive value for clinically significant CAD on a per patient basis in MPI positive symptomatic patients. In conclusion, almost half (48%) of patients with suspected CAD and an abnormal MPI study demonstrate no obstructive CAD on CCTA.

Left ventricular end-diastolic pressure affects measurement of fractional flow reserve.

BACKGROUND: Fractional flow reserve (FFR), the hyperemic ratio of distal (Pd) to proximal (Pa) coronary pressure, is used to identify the need for coronary revascularization. Changes in left ventricular end-diastolic pressure (LVEDP) might affect measurements of FFR. METHODS AND MATERIALS: LVEDP was recorded simultaneously with Pd and Pa during conventional FFR measurement as well as during additional infusion of nitroprusside. The relationship between LVEDP, Pa, and FFR was assessed using linear mixed models. RESULTS: Prospectively collected data for 528 cardiac cycles from 20 coronary arteries in 17 patients were analyzed. Baseline median Pa, Pd, FFR, and LVEDP were 73 mmHg, 49 mmHg, 0.69, and 18 mmHg, respectively. FFR<0.80 was present in 14 arteries (70%). With nitroprusside median Pa, Pd, FFR, and LVEDP were 61 mmHg, 42 mmHg, 0.68, and 12 mmHg, respectively. In a multivariable model for the entire population LVEDP was positively associated with FFR such that FFR increased by 0.008 for every 1-mmHg increase in LVEDP (beta=0.008; P<0.001), an association that was greater in obstructed arteries with FFR<0.80 (beta=0.01; P<0.001). Pa did not directly affect FFR in the multivariable model, but an interaction between LVEDP and Pa determined that LVEDP's effect on FFR is greater at lower Pa. CONCLUSIONS: LVEDP was positively associated with FFR. The association was greater in obstructive disease (FFR<0.80) and at lower Pa. These findings have implications for the use of FFR to guide revascularization in patients with heart failure. SUMMARY FOR ANNOTATED TABLE OF CONTENTS: The impact of left ventricular diastolic pressure on measurement of fractional flow reserve (FFR) is not well described. We present a hemodynamic study of the issue, concluding that increasing left ventricular diastolic pressure can increase measurements of FFR, particularly in patients with FFR<0.80 and lower blood pressure.

Comparison of lipid deposition at coronary bifurcations versus at nonbifurcation portions of coronary arteries as determined by near-infrared spectroscopy.

Atherosclerosis has been shown to develop preferentially at sites of coronary bifurcation, yet culprit lesions resulting in ST-elevation myocardial infarction do not occur more frequently at these sites. We hypothesized that these findings can be explained by similarities in intracoronary lipid and that lipid and lipid core plaque would be found with similar frequency in coronary bifurcation and nonbifurcation segments. One hundred seventy bifurcations were identified, 156 of which had comparative nonbifurcation segments proximal and/or distal to the bifurcation. We compared lipid deposition at bifurcation and nonbifurcation segments in coronary arteries using near-infrared spectroscopy (NIRS), a novel method for the in vivo detection of coronary lipid. Any NIRS signal for the presence of lipid was found with similar frequency in bifurcation and nonbifurcation segments (79% vs 74%, p = NS). Lipid core burden index, a measure of total lipid quantity indexed to segment length, was similar across bifurcation segments as well as their proximal and distal controls (lipid core burden index 66.3 ± 106, 67.1 ± 116, and 66.6 ± 104, p = NS). Lipid core plaque, identified as a high-intensity focal NIRS signal, was found in 21% of bifurcation segments, and 20% of distal nonbifurcation segments (p = NS). In conclusion, coronary bifurcations do not appear to have higher levels of intracoronary lipid or lipid core plaque than their comparative nonbifurcation regions.

Alcohol septal ablation for obstructive hypertrophic cardiomyopathy: outcomes in young, middle-aged, and elderly patients.

OBJECTIVES: We compared the efficacy and safety of alcohol septal ablation (ASA) for obstructive hypertrophic cardiomyopathy (HCM) in young, middle-aged, and elderly patients. BACKGROUND: Intersociety guidelines suggest based on limited evidence that young patients with medically refractory symptoms of obstructive HCM should undergo surgical myectomy while elderly patients may be more appropriate for ASA. METHODS: Data for 360 patients undergoing 389 ASAs were prospectively collected and retrospectively analyzed according to age. RESULTS: Young (<45 years), middle-aged (45-64 years), and elderly (≥65 years) patients comprised 28, 40, and 32% of the study population, respectively. Young patients had thicker left ventricular septal walls at baseline, and elderly patients had more comorbidity and dyspnea. Resting, mean left ventricular outflow tract gradients (LVOTGs) were similar across the age groups at baseline (62, 66, and 68 mm Hg, respectively; P = NS for all comparisons). LVOTGs and dyspnea were significantly and similarly improved in all age groups immediately after ASA and through 12 months of follow-up (P < 0.001 for before and after comparisons; P = NS for intergroup comparisons). Complication rates were similar for young and middle-aged patients but higher for elderly patients (9.1 and 6.3% vs. 20.8%, respectively; P ≤ 0.016 for elderly vs. others). Mortality rates for young and middle-aged patients were lower than for elderly patients, but the differences were not statistically significant. CONCLUSIONS: Patients undergoing ASA had significant and similar improvements in LVOTGs and symptoms regardless of age. Procedural complications were increased in elderly patients, who had numerically but not statistically significantly higher mortality rates.

Coronary CT angiography versus conventional cardiac angiography for therapeutic decision making in patients with high likelihood of coronary artery disease.

PURPOSE: To assess the efficacy of coronary computed tomographic (CT) angiography for therapeutic decision making in patients with high likelihood of coronary artery disease (CAD)-specifically the ability of coronary CT angiography to help differentiate patients without and patients with a need for revascularization and determine the appropriate revascularization procedure. MATERIALS AND METHODS: The study protocol was approved by institutional review board, with written informed consent from all patients. The study was conducted in compliance with HIPAA. One hundred eighty-five consecutive symptomatic patients (121 men; mean age, 59.4 years±9.7) with a positive single photon emission computed tomography (SPECT) myocardial perfusion study underwent coronary CT angiography and conventional cardiac angiography (hereafter, cardiac catheterization). The management strategy (conservative treatment vs revascularization) and revascularization procedure (percutaneous coronary intervention [PCI] vs coronary artery bypass graft surgery [CABG]) were prospectively selected on the basis of a combination of coronary CT angiography and SPECT. In addition, the authors calculated the accuracy, sensitivity, specificity, and negative and positive predictive values of coronary CT angiography in the detection of obstructive CAD and the selection of a revascularization strategy. Cardiac catheterization was used as the standard of reference. RESULTS: Of the 185 patients, 113 (61%) did not undergo revascularization and 42 (23%) were free of CAD. In 178 patients (96%), the same therapeutic strategy (conservative treatment vs revascularization) was chosen on the basis of coronary CT angiography and catheterization. All patients in need of revascularization were identified with coronary CT angiography. When revascularization was indicated, the same procedure (PCI vs CABG) was chosen in 66 of 72 patients (92%). CONCLUSION: In patients with high likelihood of CAD, the performance of coronary CT angiography in the differentiation of patients without and patients with a need for revascularization and the selection of a revascularization strategy was similar to that of cardiac catheterization; accordingly, coronary CT angiography has the potential to limit the number of patients without obstructive CAD who undergo cardiac catheterization and to inform decision making regarding revascularization.

Everolimus-eluting stents in interventional cardiology.

Bare metal stents have a proven safety record, but limited long-term efficacy due to in-stent restenosis. First-generation drug-eluting stents successfully countered the restenosis rate, but were hampered by concerns about their long-term safety. Second generation drug-eluting stents have combined the low restenosis rate of the first generation with improved long-term safety. We review the evolution of drug-eluting stents with a focus on the safety, efficacy, and unique characteristics of everolimus-eluting stents.

Usefulness of international normalized ratio to predict bleeding complications in patients with end-stage liver disease who undergo cardiac catheterization.

Patients with end-stage liver disease frequently require invasive cardiac procedures in preparation for liver transplantation. Because of the impaired hepatic function, these patients often have a prolonged prothrombin time and elevated international normalized ratio (INR). To determine whether an abnormal prothrombin time/INR is predictive of bleeding complications from invasive cardiac procedures, we retrospectively reviewed, for bleeding complications, the databases and case records of our series of patients with advanced cirrhosis who underwent cardiac catheterization. A total of 157 patients underwent isolated right-sided heart catheterization, and 83 underwent left-sided heart catheterization or combined left- and right-sided heart catheterization. The INR ranged from 0.93 to 2.35. No major procedure-related complications occurred. Several patients in each group required a blood transfusion for gastrointestinal bleeding but not for procedure-related bleeding. No significant change was found in the hemoglobin after right-sided or left-sided heart catheterization, and no correlation was found between the preprocedure INR and the change in postprocedure hemoglobin. When comparing patients with a normal (≤1.5) and elevated (>1.5) INR, no significant difference in hemoglobin after the procedure was found in either group. In conclusion, despite an elevated INR, patients with end-stage liver disease can safely undergo invasive cardiac procedures. An elevated INR does not predict catheterization-related bleeding complications in this patient population.

Comparison of percutaneous coronary intervention safety before and during the establishment of a transradial program at a teaching hospital.

This study sought to examine the safety of percutaneous coronary intervention (PCI) before and during de novo establishment of a transradial (TR) program at a teaching hospital. TR access remains underused in the United States, where cardiology fellowship programs continue to produce cardiologists with little TR experience. Establishment of TR programs at teaching hospitals may affect PCI safety. Starting in July 2009 a TR program was established at a teaching hospital. PCI-related data for academic years 2008 to 2009 (Y1) and 2009 to 2010 (Y2) were prospectively collected and retrospectively analyzed. Of 1,366 PCIs performed over 2 years, 0.1% in Y1 and 28.7% in Y2 were performed by TR access. No major complications were identified in 194 consecutive patients undergoing TR PCI, and combined bleeding and vascular complication rates were lower in Y2 versus Y1 (0.7% vs 2.0%, p = 0.05). Patients treated in Y2 versus Y1 and by TR versus transfemoral approach required slightly more fluoroscopy but similar contrast volumes and had similar procedural durations, lengths of stay, and predischarge mortality rates. PCI success rates were 97% in Y1, 97% in Y2, and 98% in TR cases. TR PCIs were performed by 13 cardiology fellows and 9 attending physicians, none of whom routinely performed TR PCI previously. In conclusion, de novo establishment of a TR program improved PCI safety at a teaching hospital. TR programs are likely to improve PCI safety at other teaching hospitals and should be established in all cardiology fellowship training programs.

Cardiac computed tomography and magnetic resonance imaging: the clinical use from a cardiologist's perspective.

The introduction and continued evolution of cardiac computed tomography and magnetic resonance imaging have added considerable noninvasive diagnostic insight into a wide range of frequently encountered clinical cardiology scenarios. With an increasing range of imaging modalities, and multiple methods of image acquisition in each, a detailed understanding of the clinical question at hand is often necessary to select the proper study and make optimal use of imaging data. We review common cardiac issues from a clinician's perspective, along with the unique role to be played by computed tomography and magnetic resonance imaging in each condition. This review will hopefully facilitate a strong dialogue between imagers and managing clinicians, creating a shared knowledge of both the capabilities of imaging and the management challenges that treating clinicians face.