We show that the widely used Kendall tau correlation coefficient, and the related Mallows kernel, are positive definite kernels for permutations. They offer computationally attractive alternatives to more complex kernels on the symmetric group to learn from rankings, or learn to rank. We show how to extend these kernels to partial rankings, multivariate rankings and uncertain rankings. Examples are presented on how to formulate typical problems of learning from rankings such that they can be solved with state-of-the-art kernel algorithms. We demonstrate promising results on clustering heterogeneous rank data and high-dimensional classification problems in biomedical applications.

Genome-Wide Association Study (GWAS) Higher Blood pressure Arthritides Neuropsychiatric conditions Malignancies Lower Anaemias Lipidaemias Ischaemic heart disease Genetically higher central obesity Highlights Variants in HFE and TMPRSS6 are associated with higher liver iron. There is genetic evidence that higher central obesity causes higher liver iron. Liver iron variants are not organ specific and associate with multiple diseases.

We present a universal neural vocoder based on Parallel WaveNet, with an additional conditioning network called Audio Encoder. Our universal vocoder offers real-time highquality speech synthesis on a wide range of use cases. We tested it on 43 internal speakers of diverse age and gender, speaking 20 languages in 17 unique styles, of which 7 voices and 5 styles were not exposed during training. We show that the proposed universal vocoder significantly outperforms speaker-dependent vocoders overall. We also show that the proposed vocoder outperforms several existing neural vocoder architectures in terms of naturalness and universality. These findings are consistent when we further test on more than 300 open-source voices.

Chemotherapeutic treatments are indispensable in the treatment of breast cancer. However, the emergence of multidrug-resistance, strong cell toxicity, and poor targeting selection has inhibited their clinical application. In this study, two synergistic drugs, doxorubicin (DOX) and curcumin (CUR), were co-administered to overcome multidrug resistance (MDR). Based on the characteristics of the tumor microenvironment, we developed folic acid-modified nanoparticles ((DOX + CUR)-FA-NPs) based on a star-shaped polyester (FA-TRI-CL) to enhance the tumor targeting selectivity and drug loading (DL) capacity. The (DOX + CUR)-FA-NPs displayed a characteristic spheroid morphology with an ideal diameter (186.52 nm), polydispersity index (0.024), zeta potential (–18.87 mV), and good entrapment efficiency (97.64%/78.13%, DOX/CUR) and DL (20.27%/11.29%, DOX/CUR) values. In vitro pharmacokinetic and pharmacodynamic experiments demonstrated that the (DOX + CUR)-FA-NPs were gradually released, and they displayed the highest cell apoptosis and cellular uptake in MCF-7/ADR cells. Additionally, in vivo results illustrated that (DOX + CUR)-FA-NPs not only displayed significant tumor targeting and anticancer efficacy, but also induced less pathological damage to the normal tissue. In summary, co-administered DOX and CUR appeared to reverse MDR, and this targeted combinational nanoscale delivery system could thus be a promising carrier for tumor therapies in the future.

The version presented here may differ from the published version. If citing, you are advised to consult the published version for pagination, volume/issue and date of publication Full titleProcesses underlying glycemic deterioration in type 2 diabetes: An IMI DIRECT study

No abstract

Purpose To develop microchannel-based preparation of curcumin (Cur)-loaded hybrid nanoparticles using enzyme-targeted peptides and star-shaped polycyclic lipids as carriers, and to accomplish a desirable targeted drug delivery via these nanoparticles, which could improve the bioavailability and antitumor effects of Cur. Methods The amphiphilic tri-chaintricarballylic acid-poly (ε-caprolactone)-methoxypolyethylene glycol (Tri-CL-mPEG) and the enzyme-targeted tetra-chain pentaerythritol-poly (ε-caprolactone)-polypeptide (PET-CL-P) were synthesized. The Cur-loaded enzyme-targeted hybrid nano-delivery systems (Cur-P-NPs) were prepared by using the microfluidic continuous granulation technology. The physicochemical properties, release behavior in vitro, and stability of these Cur-P-NPs were investigated. Their cytotoxicity, cellular uptake, anti-proliferative efficacy in vitro, biodistribution, and antitumor effects in vivo were also studied. Results The particle size of the prepared Cur-P-NPs was 146.1 ± 1.940 nm, polydispersity index was 0.175 ± 0.014, zeta potential was 10.1 ± 0.300 mV, encapsulation rate was 74.66 ± 0.671%, and drug loading capacity was 5.38 ± 0.316%. The stability of Cur-P-NPs was adequate, and the in vitro release rate increased with the decrease of the environmental pH. Seven days post incubation, the cumulative release values of Cur were 52.78%, 67.39%, and 98.12% at pH 7.4, pH 6.8 and pH 5.0, respectively. Cur-P-NPs exhibited better cell entry and antiproliferation efficacy against U251 cells than the Cur-solution and Cur-NPs and were safe for use. Cur-P-NPs specifically targeted tumor tissues and inhibited their growth (78.63% tumor growth inhibition rate) with low toxic effects on normal tissues. Conclusion The enzyme-targeted hybrid nanoparticles prepared in the study clearly have the tumor-targeting ability. Cur-P-NPs can effectively improve the bioavailability of Cur and have potential applications in drug delivery and tumor management.

The development of high-throughput in vitro assays to study quantitatively the toxicity of chemical compounds on genetically characterized human-derived cell lines paves the way to predictive toxicogenetics, where one would be able to predict the toxicity of any particular compound on any particular individual. In this paper we present a machine learning-based approach for that purpose, kernel multitask regression (KMR), which combines chemical characterizations of molecular compounds with genetic and transcriptomic characterizations of cell lines to predict the toxicity of a given compound on a given cell line. We demonstrate the relevance of the method on the recent DREAM8 Toxicogenetics challenge, where it ranked among the best state-of-the-art models, and discuss the importance of choosing good descriptors for cell lines and chemicals.