Characterizing heterogeneity along EMT and metabolic axes in colorectal cancer reveals underlying consensus molecular subtype-specific trends.

Colorectal cancer (CRC) is highly heterogeneous with variable survival outcomes and therapeutic vulnerabilities. A commonly used classification system in CRC is the Consensus Molecular Subtypes (CMS) based on gene expression patterns. However, how these CMS categories connect to axes of phenotypic plasticity and heterogeneity remains unclear. Here, in our analysis of CMS-specific TCGA data and 101 bulk transcriptomic datasets, we found the epithelial phenotype score to be consistently positively correlated with scores of glycolysis, OXPHOS and FAO pathways, while PD-L1 activity scores positively correlated with mesenchymal phenotype scoring, revealing possible interconnections among plasticity axes. Single-cell RNA-sequencing analysis of patient samples revealed that that CMS2 and CMS3 subtype samples were relatively more epithelial as compared to CMS1 and CMS4. CMS1 revealed two subpopulations: one close to CMS4 (more mesenchymal) and the other closer to CMS2 or CMS3 (more epithelial), indicating a partial EMT-like behavior. Consistent observations were made in single-cell analysis of metabolic axes and PD-L1 activity scores. Together, our results quantify the patterns of two functional interconnected axes of phenotypic heterogeneity - EMT and metabolic reprogramming - in a CMS-specific manner in CRC.

Dual role of CASP8AP2/FLASH in regulating epithelial-to-mesenchymal transition plasticity (EMP).

BACKGROUND: Epithelial-to-mesenchymal transition (EMT) is a developmental program that consists of the loss of epithelial features concomitant with the acquisition of mesenchymal features. Activation of EMT in cancer facilitates the acquisition of aggressive traits and cancer invasion. EMT plasticity (EMP), the dynamic transition between multiple hybrid states in which cancer cells display both epithelial and mesenchymal markers, confers survival advantages for cancer cells in constantly changing environments during metastasis. METHODS: RNAseq analysis was performed to assess genome-wide transcriptional changes in cancer cells depleted for histone regulators FLASH, NPAT, and SLBP. Quantitative PCR and Western blot were used for the detection of mRNA and protein levels. Computational analysis was performed on distinct sets of genes to determine the epithelial and mesenchymal score in cancer cells and to correlate FLASH expression with EMT markers in the CCLE collection. RESULTS: We demonstrate that loss of FLASH in cancer cells gives rise to a hybrid E/M phenotype with high epithelial scores even in the presence of TGFß, as determined by computational methods using expression of predetermined sets of epithelial and mesenchymal genes. Multiple genes involved in cell-cell junction formation are similarly specifically upregulated in FLASH-depleted cells, suggesting that FLASH acts as a repressor of the epithelial phenotype. Further, FLASH expression in cancer lines is inversely correlated with the epithelial score. Nonetheless, subsets of mesenchymal markers were distinctly up-regulated in FLASH, NPAT, or SLBP-depleted cells. CONCLUSIONS: The ZEB1low/SNAILhigh/E-cadherinhigh phenotype described in FLASH-depleted cancer cells is driving a hybrid E/M phenotype in which epithelial and mesenchymal markers coexist.

Convolutional neural networks (CNNs): concepts and applications in pharmacogenomics.

Convolutional neural networks (CNNs) have been used to extract information from various datasets of different dimensions. This approach has led to accurate interpretations in several subfields of biological research, like pharmacogenomics, addressing issues previously faced by other computational methods. With the rising attention for personalized and precision medicine, scientists and clinicians have now turned to artificial intelligence systems to provide them with solutions for therapeutics development. CNNs have already provided valuable insights into biological data transformation. Due to the rise of interest in precision and personalized medicine, in this review, we have provided a brief overview of the possibilities of implementing CNNs as an effective tool for analyzing one-dimensional biological data, such as nucleotide and protein sequences, as well as small molecular data, e.g., simplified molecular-input line-entry specification, InChI, binary fingerprints, etc., to categorize the models based on their objective and also highlight various challenges. The review is organized into specific research domains that participate in pharmacogenomics for a more comprehensive understanding. Furthermore, the future intentions of deep learning are outlined.

Recent trends in smartphone-based detection for biomedical applications: a review.

Smartphone-based imaging devices (SIDs) have shown to be versatile and have a wide range of biomedical applications. With the increasing demand for high-quality medical services, technological interventions such as portable devices that can be used in remote and resource-less conditions and have an impact on quantity and quality of care. Additionally, smartphone-based devices have shown their application in the field of teleimaging, food technology, education, etc. Depending on the application and imaging capability required, the optical arrangement of the SID varies which enables them to be used in multiple setups like bright-field, fluorescence, dark-field, and multiple arrays with certain changes in their optics and illumination. This comprehensive review discusses the numerous applications and development of SIDs towards histopathological examination, detection of bacteria and viruses, food technology, and routine diagnosis. Smartphone-based devices are complemented with deep learning methods to further increase the efficiency of the devices. Graphical Abstract.

Advances in nonlinear optical microscopy techniques for in vivo and in vitro neuroimaging.

Understanding the mechanism of the brain via optical microscopy is one of the challenges in neuroimaging, considering the complex structures. Advanced neuroimaging techniques provide a more comprehensive insight into patho-mechanisms of brain disorders, which is useful in the early diagnosis of the pathological and physiological changes associated with various neurodegenerative diseases. Recent advances in optical microscopy techniques have evolved powerful tools to overcome scattering of light and provide improved in vivo neuroimaging with sub-cellular resolution, endogenous contrast specificity, pinhole less optical sectioning capability, high penetration depth, and so on. The following article reviews the developments in various optical imaging techniques including two-photon and three-photon fluorescence, second-harmonic generation, third-harmonic generation, coherent anti-Stokes Raman scattering, and stimulated Raman scattering in neuroimaging. We have outlined the potentials and drawbacks of these techniques and their possible applications in the investigation of neurodegenerative diseases.