Access to Monomeric Lead(II) Hydrides with Remarkable Thermostability and Their Use in Catalytic Hydroboration of Carbonyl Derivatives.

We report on the remarkable stability of unprecedented, monomeric lead(II) hydrides M+[LPb(II)H]- (M[1-H]), where L = 2,6-bis(3,5-diphenylpyrrolyl)pyridine and M = (18-crown-6)potassium or ([2.2.2]-cryptand)potassium. The half-life of [K18c6][1-H] of ∼2 days in tetrahydrofuran at 25 °C is significantly longer than those reported for dimeric lead(II) hydrides supported by bulky terphenyl ligands (few hours at low temperatures), which are the only examples known for lead(II) hydride compounds. The presence of a Pb-H bond in [1-H]- was unambiguously identified by multinuclear NMR spectroscopy. Remarkably, a 1H resonance of the hydride ligand was found at δ = 41.43 ppm (1JPbH = 1312 Hz). For reactivity study, [1-H]- serves as an excellent hydroboration catalyst with high turnover numbers and turnover frequencies for several carbonyl compounds.

High-performance peptide and disulfide mapping by direct injection of intact proteins using on-line coupled UV-liquid chromatography microdroplet mass spectrometry (UVLC-MMS).

Microdroplet mass spectrometry (MMS), achieving ultra-fast enzyme digestion in the ionization source, holds great promises for innovating protein analysis. Here, in-depth protein characterization is demonstrated by direct injection of intact protein mixtures via on-line coupling MMS with capillary C4 liquid chromatography (LC) containing UV windows (UVLC-MMS) through an enzyme introduction tee. We showed complete sets of peptides of individual proteins (hemoglobin, bovine serum albumin, and ribonuclease A) in a mixture could be obtained in one injection. Such full (100%) sequence coverage, however, could not be achieved by conventional nanoLC-MS method using bottom-up approach with single enzyme. Moreover, direct injection of a chaperone α-crystalline (α-Cry) complex yielded identification of post-translational modifications including novel sites and semi-quantitative characterization including 3:1 stoichiometry ratio of αA- and αB-Cry sub-units and ∼1.4 phosphorylation/subunit on S45 (novel site) and S122 (main site) of αA-Cry, ∼0.7 phosphorylation/subunit on S19 (main site) and S45 of αB-Cry, as well as 100% acetylation on both N-termini of each subunits by matching the mass and retention time of the intact and its digested peptides. Furthermore, trifluoroacetic acid was able to be used in the mobile phase with UVLC-MMS to improve the separation of differentially reduced intact species and detectability of the droplet-digested products. This allowed us to completely map four disulfide linkages of ribonuclease A based on collision-induced dissociation of disulfide clusters, some of which would otherwise not be detected, preventing scrambling or shuffling errors arising from lengthy bulk solution digestion by the bottom-up approach. Integration of UVLC and MMS greatly improves droplet digestion efficiency and MS detection, enabling highly efficient workflow for in-depth and accurate protein characterization.

DotFAN: A Domain-Transferred Face Augmentation Net.

The performance of a convolutional neural network (CNN) based face recognition model largely relies on the richness of labeled training data. However, it is expensive to collect a training set with large variations of a face identity under different poses and illumination changes, so the diversity of within-class face images becomes a critical issue in practice. In this paper, we propose a 3D model-assisted domain-transferred face augmentation network (DotFAN) that can generate a series of variants of an input face based on the knowledge distilled from existing rich face datasets of other domains. Extending from StarGAN's architecture, DotFAN integrates with two additional subnetworks, i.e., face expert model (FEM) and face shape regressor (FSR), for latent facial code control. While FSR aims to extract face attributes, FEM is designed to capture a face identity. With their aid, DotFAN can separately learn facial feature codes and effectively generate face images of various facial attributes while keeping the identity of augmented faces unaltered. Experiments show that DotFAN is beneficial for augmenting small face datasets to improve their within-class diversity so that a better face recognition model can be learned from the augmented dataset.