In silico drug repurposing for the identification of potential candidate molecules against arboviruses infection.

Arboviral diseases caused by dengue (DENV), Zika (ZIKV) and chikungunya (CHIKV) viruses represent a major public health problem worldwide, especially in tropical areas where millions of infections occur every year. The aim of this research was to identify candidate molecules for the treatment of these diseases among the drugs currently available in the market, through in silico screening and subsequent in vitro evaluation with cell culture models of DENV and ZIKV infections. Numerous pharmaceutical compounds from antibiotics to chemotherapeutic agents presented high in silico binding affinity for the viral proteins, including ergotamine, antrafenine, natamycin, pranlukast, nilotinib, itraconazole, conivaptan and novobiocin. These five last compounds were tested in vitro, being pranlukast the one that exhibited the best antiviral activity. Further in vitro assays for this compound showed a significant inhibitory effect on DENV and ZIKV infection of human monocytic cells and human hepatocytes (Huh-7 cells) with potential abrogation of virus entry. Finally, intrinsic fluorescence analyses suggest that pranlukast may have some level of interaction with three viral proteins of DENV: envelope, capsid, and NS1. Due to its promising results, suitable accessibility in the market and reduced restrictions compared to other pharmaceuticals; the anti-asthmatic pranlukast is proposed as a drug candidate against DENV, ZIKV, and CHIKV, supporting further in vitro and in vivo assessment of the potential of this and other lead compounds that exhibited good affinity scores in silico as therapeutic agents or scaffolds for the development of new drugs against arboviral diseases.

ISeeU: Visually interpretable deep learning for mortality prediction inside the ICU.

To improve the performance of Intensive Care Units (ICUs), the field of bio-statistics has developed scores which try to predict the likelihood of negative outcomes. These help evaluate the effectiveness of treatments and clinical practice, and also help to identify patients with unexpected outcomes. However, they have been shown by several studies to offer sub-optimal performance. Alternatively, Deep Learning offers state of the art capabilities in certain prediction tasks and research suggests deep neural networks are able to outperform traditional techniques. Nevertheless, a main impediment for the adoption of Deep Learning in healthcare is its reduced interpretability, for in this field it is crucial to gain insight into the why of predictions, to assure that models are actually learning relevant features instead of spurious correlations. To address this, we propose a deep multi-scale convolutional architecture trained on the Medical Information Mart for Intensive Care III (MIMIC-III) for mortality prediction, and the use of concepts from coalitional game theory to construct visual explanations aimed to show how important these inputs are deemed by the network. Results show our model attains a ROC AUC of 0.8735 (± 0.0025) which is competitive with the state of the art of Deep Learning mortality models trained on MIMIC-III data, while remaining interpretable. Supporting code can be found at https://github.com/williamcaicedo/ISeeU.

Features of Dengue and Chikungunya Infections of Colombian Children under 24 Months of Age Admitted to the Emergency Department.

We aimed to assess clinical and laboratory differences between dengue and chikungunya in children <24 months of age in a comparative study. We collected retrospective clinical and laboratory data confirmed by NS1/IgM for dengue for 19 months (1 January 2013 to 17 August 2014). Prospective data for chikungunya confirmed by real-time polymerase chain reaction were collected for 4 months (22 September 2014-14 December 2014). Sensitivity and specificity [with 95% confidence interval (CI)] were reported for each disease diagnosis. A platelet count <150 000 cells/ml at emergency admission best characterized dengue, with a sensitivity of 67% (95% CI, 53-79) and specificity of 95% (95% CI, 82-99). The algorithm developed with classification and regression tree analysis showed a sensitivity of 93% (95% CI, 68-100) and specificity of 38% (95% CI, 9-76) to diagnose dengue. Our study provides potential differential characteristics between chikungunya and dengue in young children, especially low platelet counts.