Two-Step Machine Learning Approach for Charge-Transfer Coupling with Structurally Diverse Data.

Electronic coupling is important in determining charge-transfer rates and dynamics. Coupling strength is sensitive to both intermolecular, e.g., orientation or distance, and intramolecular degrees of freedom. Hence, it is challenging to build an accurate machine learning model to predict electronic coupling of molecular pairs, especially for those derived from the amorphous phase, for which intermolecular configurations are much more diverse than those derived from crystals. In this work, we devise a new prediction algorithm that employs two consecutive KRR models. The first model predicts molecular orbitals (MOs) from structural variation for each fragment, and coupling is further predicted by using the overlap integral included in a second model. With our two-step procedure, we achieved mean absolute errors of 0.27 meV for an ethylene dimer and 1.99 meV for a naphthalene pair, much improved accuracy amounting to 14-fold and 3-fold error reductions, respectively. In addition, MOs from the first model can also be the starting point to obtain other quantum chemical properties from atomistic structures. This approach is also compatible with a MO predictor with sufficient accuracy.

Rapid detection and quantification of Enterovirus 71 by a portable surface plasmon resonance biosensor.

This study presents the first report on a label-free detection and rapid quantification method for human enterovirus 71 (EV71) using a portable surface plasmon resonance (SPR) system. The SPR sensor instrument was configured to run on low power in a miniaturized platform to improve the device portability for a wider application both in laboratories and in the field. A color tunable organic light emitting diode in red spectrum was attached on a trapezoidal prism for the disposable light source module. The SPR signal processing using integration area under the reflectivity curve is applied for optimum signal to noise ratio (SNR) enhancement. The major capsid protein VP1 of EV71 was selected as the biomarker target in the detection study. The experimental time required for the EV71 quantification was reduced from 6 days using the conventional viral plaque assay to several minutes using the proposed method. The study results establish a detection limit of approximately 67 virus particles per milliliter (vp/ml) of EV71 in a Dulbecco's modified Eagle's medium. The VP1 detection in the portable SPR biosensor had a detection limit of approximately 4.8pg/ml in the PBS buffer. Therefore, the proposed direct EV71 viral particle quantification method can be rapidly performed in real time, with high sensitivity and less labor and without assays or fluorescence.