No abstract

In this paper, we introduce NeuralClassifier, a toolkit for neural hierarchical multi-label text classification. NeuralClassifier is designed for quick implementation of neural models for hierarchical multi-label classification task, which is more challenging and common in real-world scenarios. A salient feature is that NeuralClassifier currently provides a variety of text encoders, such as FastText, TextCNN, TextRNN, RCNN, VDCNN, DPCNN, DRNN, AttentiveConvNet and Transformer encoder, etc. It also supports other text classification scenarios, including binary-class and multiclass classification. Built on PyTorch 1 , the core operations are calculated in batch, making the toolkit efficient with the acceleration of GPU. Experiments show that models built in our toolkit achieve comparable performance with reported results in the literature.

BackgroundRecurrent laryngeal nerve (RLN) lymph node metastasis used to be shown a predictor for poor prognosis in esophageal squamous cell carcinoma. The purpose of this study was to evaluate the prognostic impact of RLN node metastasis and the number of metastatic lymph nodes in node-positive patients with squamous cell carcinoma of middle thoracic esophagus.MethodsA cohort of 235 patients who underwent curative surgery for squamous cell carcinoma of middle thoracic esophagus was investigated. The prognostic impact was evaluated by univariate and multivariate analyses.ResultsLymph node metastasis was found in 133 patients. Among them, 81 had metastatic RLN nodes, and 52 had at least one positive node but no RLN nodal involvement. The most significant difference in survival was detected between patients with metastatic lymph nodes below and above a cutoff value of six (P < 0.001). Multivariate analysis revealed that the number of metastatic lymph nodes was a significant factor associated with overall survival (P < 0.001), but RLN lymph node metastasis was not (P = 0.865).ConclusionsRLN Lymph node metastasis is not, but the number of metastatic nodes is a prognostic predictor in node-positive patients with squamous cell carcinoma of the middle thoracic esophagus.

Background: Methotrexate is the first systemic therapeutics of psoriasis. It is reported that 40% of the patients achieved a PASI75 after 12 weeks with a small dose of methotrexate (15 mg / w) treatment. So far there is not any large-scale exome sequencing been used to predict the efficacy of methotrexate in the treatment of psoriasis vulgaris. Objective: To analyze the genetic polymorphism to predict methotrexate efficacy in Chinese patients with psoriasis vulgaris. Methods: In this study, we used the whole exon high-throughput sequencing technology to detect the DNA sequence of 22 psoriasis vulgaris patients (11 responders, 11 non-responders) treated with methotrexate and captured approximately 236 variants with statistically significant in the whole exon sequencing, then in accordance with statistical differences and clinical relevance, we further selected 36 SNPs and 14 SNPs that have been reported in articles associated with the response of methotrexate. We used MassARRAY method to verify the 50 SNPs in 100 psoriatic patients treated with methotrexate. Results: We found 3 SNPs, rs216195T > C in SMG6, rs1050301G > A in IMMT, rs2285421T > C in UPK1A which might associate with the drug response of methotrexate. Conclusion:We have searched 3 new SNPs that could predict the efficacy of methotrexate in psoriasis vulgaris to some extent, providing a theoretical basis for precision medicine of methotrexate in future.

Allergic rhinitis (AR) is a common upper airway disease attributed to a variety of risk factors, such as environmental exposures and genetic susceptibility. The commonly observed comorbidity of asthma and AR in the clinic suggests the presence of shared genetic risk factors and biological mechanisms between these diseases. Interleukin (IL)-33 has been indicated to be an important factor driving asthma susceptibility and pathogenesis using both genome-wide association studies and functional studies in model animals. Although previous studies have reported the putative association of this gene with AR, evidence for the association of genetic variations of IL-33 with the disease is still missing. To examine whether variations in the IL-33 gene confer a genetic risk of AR, a total of 769 patients with AR and 769 age-and sex-matched healthy controls were recruited among Han Chinese residents in the Hubei province, and 14 single-nucleotide polymorphisms (SNPs) spanning the IL-33 gene were examined for their association with the risk of AR. The results indicated that five SNPs, which were in a moderate linkage disequilibrium and were located in the 5'-flanking region of IL-33, exhibited significant associations with the risk of AR, and these associations were additionally supported by genotypic and haplotypic analyses. Notably, three of the five IL-33 SNPs have been previously reported to exhibit genome-wide associations with asthma, and their alleles were also revealed to confer an increased risk of AR in the present study. In summary, the results of the current study suggested that certain variations in the IL-33 gene represent a potential risk for AR, and indicated a shared genetic basis between AR and asthma.