Signal transducer and activator of transcription (Stat) 3 is an oncogene constitutively activated in many cancer systems where it contributes to carcinogenesis. To develop chemical probes that selectively target Stat3, we virtually screened 920,000 small drug-like compounds by docking each into the peptide-binding pocket of the Stat3 SH2 domain, which consists of three sites—the pY-residue binding site, the +3 residue-binding site and a hydrophobic binding site, which served as a selectivity filter. Three compounds satisfied criteria of interaction analysis, competitively inhibited recombinant Stat3 binding to its immobilized pY-peptide ligand and inhibited IL-6-mediated tyrosine phosphorylation of Stat3. These compounds were used in a similarity screen of 2.47 million compounds, which identified 3 more compounds with similar activities. Examination of the 6 active compounds for the ability to inhibit IFN-γ-mediated Stat1 phosphorylation revealed that 5 of 6 were selective for Stat3. Molecular modeling of the SH2 domains of Stat3 and Stat1 bound to compound revealed that compound interaction with the hydrophobic binding site was the basis for selectivity. All 5 selective compounds inhibited nuclear-to-cytoplasmic translocation of Stat3, while 3 of 5 compounds induced apoptosis preferentially of breast cancer cell lines with constitutive Stat3 activation. Thus, virtual ligand screening of compound libraries that targeted the Stat3 pY-peptide binding pocket identified for the first time 3 lead compounds that competitively inhibited Stat3 binding to its pY-peptide ligand; these compounds were selective for Stat3 vs. Stat1 and induced apoptosis preferentially of breast cancer cells lines with constitutively activated Stat3.

Human awareness plays an important role in the spread of infectious diseases and the control of propagation patterns. The dynamic process with human awareness is called awareness cascade, during which individuals exhibit herd-like behavior because they are making decisions based on the actions of other individuals [Borge-Holthoefer et al., J. Complex Networks 1, 3 (2013)]. In this paper, to investigate the epidemic spreading with awareness cascade, we propose a local awareness controlled contagion spreading model on multiplex networks. By theoretical analysis using a microscopic Markov chain approach and numerical simulations, we find the emergence of an abrupt transition of epidemic threshold β(c) with the local awareness ratio α approximating 0.5, which induces two-stage effects on epidemic threshold and the final epidemic size. These findings indicate that the increase of α can accelerate the outbreak of epidemics. Furthermore, a simple 1D lattice model is investigated to illustrate the two-stage-like sharp transition at α(c)≈0.5. The results can give us a better understanding of why some epidemics cannot break out in reality and also provide a potential access to suppressing and controlling the awareness cascading systems.

Renin-angiotensin (RAS) genes, a group of promising candidate genes involved in essential hypertension (EH), play a key role in blood pressure regulation. Recently, a series of novel RAS gene polymorphisms were reported, which significantly influence the rate of the gene transcription. This study was designed to explore the association between the RAS gene polymorphisms and EH in a remote countryside population. We examined six polymorphisms in the main component genes of RAS: angiotensin-converting enzyme (ACE) (I/D), angiotensinogen (AGT) (A-6G, A-20C, G-217A and T174 M) and angiotensin type 1 receptor (AT1R) (A1166C). Six polymorphisms were genotyped by gene chip technology. Association studies were performed in 220 EH patients and 235 normotensives.Our results revealed that AGT A-6G, T174 M and ACE-I/D were significantly associated with EH (AGT A-6G: AG þ GG vs AA; OR ¼ 1.36; 95% CI ¼ 1.04-1.77. T174M: CT þ TT vs CC; OR ¼ 1.45; 95% CI ¼ 1.15-1.90. ACE I/D: ID þ DD vs II; OR ¼ 1.171; 95% CI ¼ 1.00-1.37). Moreover the logistic regression analysis suggested that the haplotype of AGT À6A, 174C, À217G and À20A might decrease the risk of EH (OR ¼ 0.64; 95% CI ¼ 0.49-0.83), after adjusting the confounding factors of gender, age and BMI. In conclusion, the AGT A-6G, T174 M and ACE I/D polymorphisms are associated with EH and the AGT haplotype À6A, 174C, À217G and À20A decrease the risk of EH in the southern Chinese population.

In this paper, we propose to study the problem of COURT VIEW GENeration from the fact description in a criminal case. The task aims to improve the interpretability of charge prediction systems and help automatic legal document generation. We formulate this task as a text-to-text natural language generation (NLG) problem. Sequenceto-sequence model has achieved cutting-edge performances in many NLG tasks. However, due to the non-distinctions of fact descriptions, it is hard for Seq2Seq model to generate charge-discriminative court views. In this work, we explore charge labels to tackle this issue. We propose a label-conditioned Seq2Seq model with attention for this problem, to decode court views conditioned on encoded charge labels. Experimental results show the effectiveness of our method.

BC characteristics in Central China displayed representative patterns of Mainland China, while showed distinct patterns from Chinese patients in other developed areas and USA.