Asthma is a very heterozygous disease, divided in subtypes, such as eosinophilic and neutrophilic asthma. Phenotyping and endotyping of patients, especially patients with severe asthma who are refractory to standard treatment, are crucial in asthma management and are based on a combination of clinical and biological features. Nevertheless, the quest remains to find better biomarkers that distinguish asthma subtypes in a more clear and objective manner and to find new therapeutic targets to treat people with therapy-resistant asthma. In the past, research to identify asthma subtypes mainly focused on expression profiles of protein-coding genes. However, advances in RNA-sequencing technologies and the discovery of non-coding RNAs as important post-transcriptional regulators have provided an entire new field of research opportunities in asthma. This review focusses on long non-coding RNAs (lncRNAs) in asthma; these are non-coding RNAs with a length of more than 200 nucleotides. Many lncRNAs are differentially expressed in asthma, and several have been associated with asthma severity or inflammatory phenotype. Moreover, in vivo and in vitro functional studies have identified the mechanisms of action of specific lncRNAs. Although lncRNAs remain not widely studied in asthma, the current studies show the potential of lncRNAs as biomarkers and therapeutic targets as well as the need for further research.

The dominant sequence transduction models are based on complex recurrent or convolutional neural networks that include an encoder and a decoder. The best performing models also connect the encoder and decoder through an attention mechanism. We propose a new simple network architecture, the Transformer, based solely on attention mechanisms, dispensing with recurrence and convolutions entirely. Experiments on two machine translation tasks show these models to be superior in quality while being more parallelizable and requiring significantly less time to train. Our model achieves 28.4 BLEU on the WMT 2014 Englishto-German translation task, improving over the existing best results, including ensembles, by over 2 BLEU. On the WMT 2014 English-to-French translation task, our model establishes a new single-model state-of-the-art BLEU score of 41.0 after training for 3.5 days on eight GPUs, a small fraction of the training costs of the best models from the literature. * Equal contribution. Listing order is random. Jakob proposed replacing RNNs with self-attention and started the effort to evaluate this idea. Ashish, with Illia, designed and implemented the first Transformer models and has been crucially involved in every aspect of this work. Noam proposed scaled dot-product attention, multi-head attention and the parameter-free position representation and became the other person involved in nearly every detail. Niki designed, implemented, tuned and evaluated countless model variants in our original codebase and tensor2tensor. Llion also experimented with novel model variants, was responsible for our initial codebase, and efficient inference and visualizations. Lukasz and Aidan spent countless long days designing various parts of and implementing tensor2tensor, replacing our earlier codebase, greatly improving results and massively accelerating our research.† Work performed while at Google Brain. ‡ Work performed while at Google Research.

No abstract

In vitro cell culture experiments are widely used to study cellular behavior in most biological research fields. Except for suspension cells, most human cell types are cultured as adherent monolayers on a plastic surface. While technically convenient, monolayer cultures can suffer from limitations in terms of physiological relevance, as their resemblance to complex in vivo tissue structures is limited. To address these limitations, three-dimensional (3D) cell culture systems have gained increased interest as they mimic key structural and functional properties of their in vivo tissue counterparts. Nevertheless, protocols established on monolayer cell cultures may require adjustments if they are to be applied to 3D cell cultures. As gene expression quantification is an essential part of many in vitro experiments, we evaluated and optimized a direct cell lysis, reverse transcription and qPCR protocol applicable for 3D cell cultures. The newly developed protocol wherein gene expression is determined directly from crude cell lysates showed improved cell lysis compared to the standard protocol, accurate gene expression quantification, hereby avoiding time-consuming cell harvesting and RNA extraction.