Warfarin dosing remains challenging due to narrow therapeutic index and highly individual variability. Incorrect warfarin dosing is associated with devastating adverse events. Remarkable efforts have been made to develop the machine learning based warfarin dosing algorithms incorporating clinical factors and genetic variants such as polymorphisms in CYP2C9 and VKORC1. The most widely validated pharmacogenetic algorithm is the IWPC algorithm based on multivariate linear regression (MLR). However, with only a single algorithm, the prediction performance may reach an upper limit even with optimal parameters. Here, we present novel algorithms using stacked generalization frameworks to estimate the warfarin dose, within which different types of machine learning algorithms function together through a meta-machine learning model to maximize the prediction accuracy. Compared to the IWPC-derived MLR algorithm, Stack 1 and 2 based on stacked generalization frameworks performed significantly better overall. Subgroup analysis revealed that the mean of the percentage of patients whose predicted dose of warfarin within 20% of the actual stable therapeutic dose (mean percentage within 20%) for Stack 1 was improved by 12.7% (from 42.47% to 47.86%) in Asians and by 13.5% (from 22.08% to 25.05%) in the low-dose group compared to that for MLR, respectively. These data suggest that our algorithms would especially benefit patients requiring low warfarin maintenance dose, as subtle changes in warfarin dose could lead to adverse clinical events (thrombosis or bleeding) in patients with low dose. Our study offers novel pharmacogenetic algorithms for clinical trials and practice.
14 August 2019 in Journal of the American Chemical Society
Spin based properties, applications, and devices are typically related to inorganic ferromagnetic materials. The development of organic materials for spintronic applications has long been encumbered by its reliance on ferromagnetic electrodes for polarized spin injection. The discovery of the chirality-induced spin selectivity (CISS) effect, in which chiral organic molecules serve as spin filters, defines a marked departure from this paradigm because it exploits soft materials, operates at ambient temperature, and eliminates the need for a magnetic electrode. To date, the CISS effect has been explored exclusively in molecular insulators. Here we combine chiral molecules, which serve as spin filters, with molecular wires that despite not being chiral, function to preserve spin polarization. Self-assembled monolayers (SAMs) of right-handed helical (l-proline)8 (Pro8) and corresponding peptides, N-terminal conjugated to (porphinato)zinc or meso-to-meso ethyne-bridged (porphinato)zinc structures (Pro8PZnn), were interrogated via magnetic conducting atomic force microscopy (mC-AFM), spin-dependent electrochemistry, and spin Hall devices that measure the spin polarizability that accompanies the charge polarization. These data show that chiral molecules are not required to transmit spin-polarized currents made possible by the CISS mechanism. Measured Hall voltages for Pro8PZn1–3 substantially exceed that determined for the Pro8 control and increase dramatically as the conjugation length of the achiral PZnn component increases; mC-AFM data underscore that measured spin selectivities increase with an increasing Pro8PZn1–3 N-terminal conjugation. Because of these effects, spin-dependent electrochemical data demonstrate that spin-polarized currents, which trace their genesis to the chiral Pro8 moiety, propagate with no apparent dephasing over the augmented Pro8PZnn length scales, showing that spin currents may be transmitted over molecular distances that greatly exceed the length of the chiral moiety that makes possible the CISS effect.
12 April 2018 in The Journal of Physical Chemistry C
Metalloproteins are a family of proteins characterized by metal ion binding, whereby the presence of these ions confers key catalytic and ligand-binding properties. Due to their ubiquity among biological systems, researchers have made immense efforts to predict the structural and functional roles of metalloproteins. Ultimately, having a comprehensive understanding of metalloproteins will lead to tangible applications, such as designing potent inhibitors in drug discovery. Recently, there has been an acceleration in the number of studies applying machine learning to predict metalloprotein properties, primarily driven by the advent of more sophisticated machine learning algorithms. This review covers how machine learning tools have consolidated and expanded our comprehension of various aspects of metalloproteins (structure, function, stability, ligand-binding interactions, and inhibitors). Future avenues of exploration are also discussed.
An investigation of spin and conformational dynamics in a series of symmetric Cu–Cu porphyrin dimer solutions is presented using electron paramagnetic resonance (EPR) spectroscopy. Previous spectral simulations focused on the isotropic exchange interaction (Javg) between the Cu centers. In this work, an additional line broadening parameter (Jmod) is explored in detail via variable temperature X-band EPR in liquid solution for several different structures. The Jmod phenomenon is due to fluctuations in the spin exchange interaction caused by conformational motion of the porphyrin planes. The Jmod parameter scales with the inverse of the rotational barriers that determine the Boltzmann-weighted torsional angle distribution between neighboring porphyrin planes. Arrhenius plots allow for extraction of the activation energies for rotation, which are 5.77, 2.84, and 5.31 kJ/mol for ethyne-bridged (porphinato)copper(II)-(porphinato)copper(II), butadiyne-bridged (porphinato)copper(II)-(porphinato)copper(II), and ethyne-bridged (porphinato)copper(II)-(porphinato)zinc(II)-(porphinato)copper(II) complexes, respectively. DFT calculations of these torsional barriers match well with the experimental results. This is the first report of a Jmod analysis within a highly anisotropic hyperfine field and demonstrates the utility of the theory for extraction of dynamic information.
14 August 2020 in The Journal of Physical Chemistry A