Background Lung adenocarcinoma, the most common type of lung cancer, has a high level of morphologic heterogeneity and is composed of tumor cells of multiple histological subtypes. It has been reported that immune cell infiltration significantly impacts clinical outcomes of patients with lung adenocarcinoma. However, it is unclear whether histologic subtyping can reflect the tumor immune microenvironment, and whether histologic subtyping can be applied for therapeutic stratification of the current standard of care. Methods We inferred immune cell infiltration levels using a histological subtype-specific gene expression dataset. From differential gene expression analysis between different histological subtypes, we developed two gene signatures to computationally determine the relative abundance of lepidic and solid components (denoted as the L-score and S-score, respectively) in lung adenocarcinoma samples. These signatures enabled us to investigate the relationship between histological composition and clinical outcomes in lung adenocarcinoma using previously published datasets. Results We found dramatic immunological differences among histological subtypes. Differential gene expression analysis showed that the lepidic and solid subtypes could be differentiated based on their gene expression patterns while the other subtypes shared similar gene expression patterns. Our results indicated that higher L-scores were associated with prolonged survival, and higher S-scores were associated with shortened survival. L-scores and S-scores were also correlated with global genomic features such as tumor mutation burdens and driver genomic events. Interestingly, we observed significantly decreased L-scores and increased S-scores in lung adenocarcinoma samples with EGFR gene amplification but not in samples with EGFR gene mutations. In lung cancer cell lines, we observed significant correlations between L-scores and cell sensitivity to a number of targeted drugs including EGFR inhibitors. Moreover, lung cancer patients with higher L-scores were more likely to benefit from immune checkpoint blockade therapy. Conclusions Our findings provided further insights into evaluating histology composition in lung adenocarcinoma. The established signatures reflected that lepidic and solid subtypes in lung adenocarcinoma would be associated with prognosis, genomic features, and responses to targeted therapy and immunotherapy. The signatures therefore suggested potential clinical translation in predicting patient survival and treatment responses. In addition, our framework can be applied to other types of cancer with heterogeneous histological subtypes.

Edited by John M. DenuThe packaging of genomic DNA into nucleosomes creates a barrier to transcription that can be relieved through ATP-dependent chromatin remodeling via complexes such as the switch-sucrose non-fermentable (SWI-SNF) chromatin remodeling complex. The SWI-SNF complex remodels chromatin via conformational or positional changes of nucleosomes, thereby altering the access of transcriptional machinery to target genes. The SWI-SNF complex has limited ability to bind to sequencespecific elements, and, therefore, its recruitment to target loci is believed to require interaction with DNA-associated transcription factors. The Cdx family of homeodomain transcript ion factors (Cdx1, Cdx2, and Cdx4) are essential for a number of developmental programs in the mouse. Cdx1 and Cdx2 also regulate intestinal homeostasis throughout life. Although a number of Cdx target genes have been identified, the basis by which Cdx members impact their transcription is poorly understood. We have found that Cdx members interact with the SWI-SNF complex and make direct contact with Brg1, a catalytic member of SWI-SNF. Both Cdx2 and Brg1 co-occupy a number of Cdx target genes, and both factors are necessary for transcriptional regulation of such targets. Finally, Cdx2 and Brg1 occupancy occurs coincident with chromatin remodeling at some of these loci. Taken together, our findings suggest that Cdx transcription factors regulate target gene expression, in part, through recruitment of Brg1-associated SWI-SNF chromatin remodeling activity.The Cdx genes are vertebrate orthologs of Drosophila caudal and encode homeodomain transcriptional factors. In the mouse, the three members of this family (Cdx1, Cdx2, and Cdx4) are co-expressed in the caudal embryo in all three germ layers commencing at mid-gastrulation and play overlapping roles in a number of developmental programs, including axial elongation, endoderm specification, and anterior-posterior vertebral patterning (1-6). Both Cdx1 and Cdx2 are also expressed in the intestinal epithelium throughout life, where they play critical roles in intestinal homeostasis (7-11) and can function as tumor suppressors (12-15). Although Cdx members play critical roles in governing gene expression during development and in the adult intestine, little is known about the mechanisms by which Cdx members regulate target gene transcription.The packaging of DNA into nucleosomes creates a barrier to transcription by obstructing access of the basal transcription machinery as well as modulating access of other transcriptional regulators to their DNA binding motifs (16 -19). Alteration of the chromatin structure by ATP-dependent remodeling complexes can alleviate these constraints, and such complexes play important roles in the transcriptional regulation of many eukaryotic genes. Such complexes include the switch-sucrose non-fermentable (SWI-SNF) 2 chromatin-remodeling complex (16, 20 -22), which remodels the chromatin structure via conformational or positional changes of nucleosomes (23,24). Because the SWI-...

From the end of 2019, one of the most serious and largest spread pandemics occurred in Wuhan (China) named Coronavirus (COVID-19). As reported by the World Health Organization, there are currently more than 100 million infectious cases with an average mortality rate of about five percent all over the world. To avoid serious consequences on people’s lives and the economy, policies and actions need to be suitably made in time. To do that, the authorities need to know the future trend in the development process of this pandemic. This is the reason why forecasting models play an important role in controlling the pandemic situation. However, the behavior of this pandemic is extremely complicated and difficult to be analyzed, so that an effective model is not only considered on accurate forecasting results but also the explainable capability for human experts to take action pro-actively. With the recent advancement of Artificial Intelligence (AI) techniques, the emerging Deep Learning (DL) models have been proving highly effective when forecasting this pandemic future from the huge historical data. However, the main weakness of DL models is lacking the explanation capabilities. To overcome this limitation, we introduce a novel combination of the Susceptible-Infectious-Recovered-Deceased (SIRD) compartmental model and Variational Autoencoder (VAE) neural network known as BeCaked. With pandemic data provided by the Johns Hopkins University Center for Systems Science and Engineering, our model achieves 0.98 $$R^2$$ R 2 and 0.012 MAPE at world level with 31-step forecast and up to 0.99 $$R^2$$ R 2 and 0.0026 MAPE at country level with 15-step forecast on predicting daily infectious cases. Not only enjoying high accuracy, but BeCaked also offers useful justifications for its results based on the parameters of the SIRD model. Therefore, BeCaked can be used as a reference for authorities or medical experts to make on time right decisions.