Word embeddings are a popular approach to unsupervised learning of word relationships that are widely used in natural language processing. In this article, we present a new set of embeddings for medical concepts learned using an extremely large collection of multimodal medical data. Leaning on recent theoretical insights, we demonstrate how an insurance claims database of 60 million members, a collection of 20 million clinical notes, and 1.7 million full text biomedical journal articles can be combined to embed concepts into a common space, resulting in the largest ever set of embeddings for 108,477 medical concepts. To evaluate our approach, we present a new benchmark methodology based on statistical power specifically designed to test embeddings of medical concepts. Our approach, called cui2vec, attains state-of-the-art performance relative to previous methods in most instances. Finally, we provide a downloadable set of pre-trained embeddings for other researchers to use, as well as an online tool for interactive exploration of the cui2vec embeddings.

A homogeneous cobalt-catalyzed β-alkylation of secondary alcohols with primary alcohols to selectively synthesize ketones via acceptorless dehydrogenative coupling is reported for the first time. Notably, this transformation is environmentally benign and atom economical with water and hydrogen gas as the only byproducts.

A novel catalytic system with a tripodal cobalt complex is developed for efficiently converting primary alcohols to esters. KO Bu is found essential to the transformation. A preliminary mechanistic study suggests a plausible reaction route that involves an initial Co-catalyzed dehydrogenation of alcohol to aldehyde, followed by a Tishchenko-type pathway to ester mediated by KO Bu.

In this paper, we demonstrate for the first time that poly(m-phenylenediamine) (PMPD) nanospheres and nanorods can be selectively synthesized via chemical oxidation polymerization of m-phenylenediamine (MPD) monomers using ammonium persulfate (APS) as an oxidant at room temperature. It suggests that the pH value plays a critical role in controlling the the morphology of the nanostructures and fast polymerization rate favors the anisotropic growth of PMPD under homogeneous nucleation condition. We further demonstrate that such PMPD nanostructures can be used as an effective fluorescent sensing platform for multiplex nucleic acid detection. A detection limit as low as 50 pM and a high selectivity down to single-base mismatch could be achieved. The fluorescence quenching is attributed to photoinduced electron transfer from nitrogen atom in PMPD to excited fluorophore. Most importantly, the successful use of this sensing platform in human blood serum system is also demonstrated.

Traditional methods for identifying food-borne pathogens are time-consuming and laborious, so it is necessary to develop innovative methods for the rapid identification of food-borne pathogens. Here, we report the development of silicon-based optical thin-film biosensor chips for sensitive detection of 11 food-borne pathogens. Briefly, aldehyde-labeled probes were arrayed and covalently attached to a hydrazine-derivatized chip surface, and then, biotinylated polymerase chain reaction (PCR) amplicons were hybridized with the probes. After washing and brief incubation with an antibiotin immunoglobulin G-horseradish peroxidase conjugate and a precipitable horseradish peroxidase substrate, biotinylated chains bound to the probes were visualized as a color change on the chip surface (gold to blue/purple). Highly sensitive and accurate examination of PCR fragment targets can be completed within 30 min. This assay is extremely robust, sensitive, specific, and economical and can be adapted to different throughputs. Thus, a rapid, sensitive, and reliable technique for detecting 11 food-borne pathogens was successfully developed.