Deep feature extraction of single-cell transcriptomes by generative adversarial network.

MOTIVATION: Single-cell RNA-sequencing (scRNA-seq) offers the opportunity to dissect heterogeneous cellular compositions and interrogate the cell-type-specific gene expression patterns across diverse conditions. However, batch effects such as laboratory conditions and individual-variability hinder their usage in cross-condition designs. RESULTS: Here, we present a single-cell Generative Adversarial Network (scGAN) to simultaneously acquire patterns from raw data while minimizing the confounding effect driven by technical artifacts or other factors inherent to the data. Specifically, scGAN models the data likelihood of the raw scRNA-seq counts by projecting each cell onto a latent embedding. Meanwhile, scGAN attempts to minimize the correlation between the latent embeddings and the batch labels across all cells. We demonstrate scGAN on three public scRNA-seq datasets and show that our method confers superior performance over the state-of-the-art methods in forming clusters of known cell types and identifying known psychiatric genes that are associated with major depressive disorder. AVAILABILITYAND IMPLEMENTATION: The scGAN code and the information for the public scRNA-seq datasets are available at https://github.com/li-lab-mcgill/singlecell-deepfeature. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

Effect of caffeic acid on feeding, α-amylase and protease activities and allatostatin-A content of Egyptian cotton leafworm, Spodoptera littoralis (Lepidoptera: Noctuidae).

Efficiency of caffeic acid (CA) on Spodoptera littoralis was investigated. CA was mixed with artificial diet, and feeding indices and allatostatin-A (AST-A) content of the midgut were measured 10 days later. α-Amylase and protease activities were evaluated for 10 days. CA significantly decreased feeding indices. Feeding on an artificial diet containing CA decreased protease and α-amylase activities in the midgut. The incubation of the dissected midgut with AST-A increased α-amylase and protease activities. The injection of AST-A into the hemolymph of larvae also increased protease and α-amylase activities. Competitive ELISA and immunohistochemistry results showed that starvation decreased the AST-A titer and AST-A immunoreactivity (AST-A-ir) cells in the midgut whereas refeeding increased it. Here, for the first time we showed that feeding on an artificial diet containing CA also caused the AST-A level to decrease in the midgut, which itself caused α-amylase and protease activities to decrease.

Evaluation of Teaching Methods in Mass CPCR Training in Different Groups of the Society, an Observational Study.

To determine the efficacy of different methods of cardiopulmonary resuscitation (CPCR) training in 3 different groups of the society. In a prospective and observational study of 2000 individuals in 3 different groups including G1, G2, and G3 4 different protocols of CPCR training were applied and their efficacy was compared between the groups. Also, 12 months after the study course, 460 participants from 3 groups were asked to take apart in a theoretical and practical examination to evaluate the long-term efficacy of the 4 protocols. Among 2000 individuals took a parted in the study, 950 (47.5%) were G1, 600 (30%) were G2, and 450 (22.5%) were G3. G1 in 4 groups were 2.37 and 2.65 times more successful in pretest theoretical and 2.61 and 18.20 times more successful in practical examinations compared with G2 and G3 and gained highest improvement in CPCR skills. Other groups also showed significantly improved CPCR skills. Comparison of different methods of CPCR learning showed that the workshop using interactive lecture as well as human model, educational film, and reference CPCR book has the highest efficacy in all groups. This protocol of CPCR training showed more efficacy in long-term postdelayed evaluation. On the contrary, medical students had better long-term outcomes from the course. Although G1 and G2 obtained better results in learning CPCR skills, in G3 also the theoretical and practical knowledge were improved significantly. This course increased confidence for doing CPCR in all groups of the study. Considering that the most of the bystanders at emergency states are general population, training this group of the society and increasing their confidence about performing CPCR can be so effective and lifesaving at emergency states. (Clinical trial. Gov registration: NCT02120573).