With the rapid development of image scanning techniques and visualization software, whole slide imaging (WSI) is becoming a routine diagnostic method. Accelerating clinical diagnosis from pathology ...images and automating image analysis efficiently and accurately remain significant challenges. Recently, deep learning algorithms have shown great promise in pathology image analysis, such as in tumor region identification, metastasis detection, and patient prognosis. Many machine learning algorithms, including convolutional neural networks, have been proposed to automatically segment pathology images. Among these algorithms, segmentation deep learning algorithms such as fully convolutional networks stand out for their accuracy, computational efficiency, and generalizability. Thus, deep learning–based pathology image segmentation has become an important tool in WSI analysis. In this review, the pathology image segmentation process using deep learning algorithms is described in detail. The goals are to provide quick guidance for implementing deep learning into pathology image analysis and to provide some potential ways of further improving segmentation performance. Although there have been previous reviews on using machine learning methods in digital pathology image analysis, this is the first in-depth review of the applications of deep learning algorithms for segmentation in WSI analysis.
Aberrant signaling through the class I phosphatidylinositol 3-kinase (PI3K)—Akt axis is frequent in human cancer. Here, we show that Beclin 1, an essential autophagy and tumor suppressor protein, is ...a target of the protein kinase Akt. Expression of a Beclin 1 mutant resistant to Akt-mediated phosphorylation increased autophagy, reduced anchorage-independent growth, and inhibited Akt-driven tumorigenesis. Akt-mediated phosphorylation of Beclin 1 enhanced its interactions with 14-3-3 and vimentin intermediate filament proteins, and vimentin depletion increased autophagy and inhibited Akt-driven transformation. Thus, Akt-mediated phosphorylation of Beclin 1 functions in autophagy inhibition, oncogenesis, and the formation of an autophagy-inhibitory Beclin 1/14-3-3/vimentin intermediate filament complex. These findings have broad implications for understanding the role of Akt signaling and intermediate filament proteins in autophagy and cancer.
Gene regulatory networks reveal how genes work together to carry out their biological functions. Reconstructions of gene networks from gene expression data greatly facilitate our understanding of ...underlying biological mechanisms and provide new opportunities for biomarker and drug discoveries. In gene networks, a gene that has many interactions with other genes is called a hub gene, which usually plays an essential role in gene regulation and biological processes. In this study, we developed a method for reconstructing gene networks using a partial correlation-based approach that incorporates prior information about hub genes. Through simulation studies and two real-data examples, we compare the performance in estimating the network structures between the existing methods and the proposed method.
In simulation studies, we show that the proposed strategy reduces errors in estimating network structures compared to the existing methods. When applied to Escherichia coli, the regulation network constructed by our proposed ESPACE method is more consistent with current biological knowledge than the SPACE method. Furthermore, application of the proposed method in lung cancer has identified hub genes whose mRNA expression predicts cancer progress and patient response to treatment.
We have demonstrated that incorporating hub gene information in estimating network structures can improve the performance of the existing methods.
Summary
Synonymous codons are not used with equal frequencies in most genomes. Codon usage has been proposed to play a role in regulating translation kinetics and co‐translational protein folding. ...The relationship between codon usage and protein structures and the in vivo role of codon usage in eukaryotic protein folding is not clear. Here, we show that there is a strong codon usage bias in the filamentous fungus Neurospora. Importantly, we found genome‐wide correlations between codon choices and predicted protein secondary structures: Nonoptimal codons are preferentially used in intrinsically disordered regions, and more optimal codons are used in structured domains. The functional importance of such correlations in vivo was confirmed by structure‐based codon manipulation of codons in the Neurospora circadian clock gene frequency (frq). The codon optimization of the predicted disordered, but not well‐structured regions of FRQ impairs clock function and altered FRQ structures. Furthermore, the correlations between codon usage and protein disorder tendency are conserved in other eukaryotes. Together, these results suggest that codon choices and protein structures co‐evolve to ensure proper protein folding in eukaryotic organisms.
Synonymous codons are not used with equal frequencies in almost eukaryotic and prokaryotic genomes. We discover that frequently used codons are preferentially used in regions of mRNAs that encode for well‐structured protein regions whereas the predicted disordered regions of proteins are preferentially encoded by non‐preferred codons. These results suggest a “code” within the genetic codons that is adapted to protein structure to ensure proper co‐translational protein folding during mRNA translation process.
Beclin 1, an autophagy and haploinsufficient tumor-suppressor protein, is frequently monoallelically deleted in breast and ovarian cancers. However, the precise mechanisms by which Beclin 1 inhibits ...tumor growth remain largely unknown. To address this question, we performed a genome-wide CRISPR/Cas9 screen in MCF7 breast cancer cells to identify genes whose loss of function reverse Beclin 1-dependent inhibition of cellular proliferation. Small guide RNAs targeting
and
, tumor-suppressor genes that encode cadherin/catenin complex members E-cadherin and alpha-catenin, respectively, were highly enriched in the screen. CRISPR/Cas9-mediated knockout of
or
reversed Beclin 1-dependent suppression of breast cancer cell proliferation and anchorage-independent growth. Moreover, deletion of
or
inhibited the tumor-suppressor effects of Beclin 1 in breast cancer xenografts. Enforced Beclin 1 expression in MCF7 cells and tumor xenografts increased cell surface localization of E-cadherin and decreased expression of mesenchymal markers and beta-catenin/Wnt target genes. Furthermore, CRISPR/Cas9-mediated knockout of
and the autophagy class III phosphatidylinositol kinase complex 2 (PI3KC3-C2) gene,
, but not PI3KC3-C1-specific
or other autophagy genes
,
, or
, resulted in decreased E-cadherin plasma membrane and increased cytoplasmic E-cadherin localization. Taken together, these data reveal previously unrecognized cooperation between Beclin 1 and E-cadherin-mediated tumor suppression in breast cancer cells.
Pathological examination of histopathological slides is a routine clinical procedure for lung cancer diagnosis and prognosis. Although the classification of lung cancer has been updated to become ...more specific, only a small subset of the total morphological features are taken into consideration. The vast majority of the detailed morphological features of tumor tissues, particularly tumor cells’ surrounding microenvironment, are not fully analyzed. The heterogeneity of tumor cells and close interactions between tumor cells and their microenvironments are closely related to tumor development and progression. The goal of this study is to develop morphological feature–based prediction models for the prognosis of patients with lung cancer.
We developed objective and quantitative computational approaches to analyze the morphological features of pathological images for patients with NSCLC. Tissue pathological images were analyzed for 523 patients with adenocarcinoma (ADC) and 511 patients with squamous cell carcinoma (SCC) from The Cancer Genome Atlas lung cancer cohorts. The features extracted from the pathological images were used to develop statistical models that predict patients’ survival outcomes in ADC and SCC, respectively.
We extracted 943 morphological features from pathological images of hematoxylin and eosin–stained tissue and identified morphological features that are significantly associated with prognosis in ADC and SCC, respectively. Statistical models based on these extracted features stratified NSCLC patients into high-risk and low-risk groups. The models were developed from training sets and validated in independent testing sets: a predicted high-risk group versus a predicted low-risk group (for patients with ADC: hazard ratio = 2.34, 95% confidence interval: 1.12–4.91, p = 0.024; for patients with SCC: hazard ratio = 2.22, 95% confidence interval: 1.15–4.27, p = 0.017) after adjustment for age, sex, smoking status, and pathologic tumor stage.
The results suggest that the quantitative morphological features of tumor pathological images predict prognosis in patients with lung cancer.
Exercise has beneficial effects on human health, including protection against metabolic disorders such as diabetes. However, the cellular mechanisms underlying these effects are incompletely ...understood. The lysosomal degradation pathway, autophagy, is an intracellular recycling system that functions during basal conditions in organelle and protein quality control. During stress, increased levels of autophagy permit cells to adapt to changing nutritional and energy demands through protein catabolism. Moreover, in animal models, autophagy protects against diseases such as cancer, neurodegenerative disorders, infections, inflammatory diseases, ageing and insulin resistance. Here we show that acute exercise induces autophagy in skeletal and cardiac muscle of fed mice. To investigate the role of exercise-mediated autophagy in vivo, we generated mutant mice that show normal levels of basal autophagy but are deficient in stimulus (exercise- or starvation)-induced autophagy. These mice (termed BCL2 AAA mice) contain knock-in mutations in BCL2 phosphorylation sites (Thr69Ala, Ser70Ala and Ser84Ala) that prevent stimulus-induced disruption of the BCL2-beclin-1 complex and autophagy activation. BCL2 AAA mice show decreased endurance and altered glucose metabolism during acute exercise, as well as impaired chronic exercise-mediated protection against high-fat-diet-induced glucose intolerance. Thus, exercise induces autophagy, BCL2 is a crucial regulator of exercise- (and starvation)-induced autophagy in vivo, and autophagy induction may contribute to the beneficial metabolic effects of exercise.
Fibroblast growth factor-21 (FGF21) is a hormone secreted by the liver during fasting that elicits diverse aspects of the adaptive starvation response. Among its effects, FGF21 induces hepatic fatty ...acid oxidation and ketogenesis, increases insulin sensitivity, blocks somatic growth and causes bone loss. Here we show that transgenic overexpression of FGF21 markedly extends lifespan in mice without reducing food intake or affecting markers of NAD+ metabolism or AMP kinase and mTOR signaling. Transcriptomic analysis suggests that FGF21 acts primarily by blunting the growth hormone/insulin-like growth factor-1 signaling pathway in liver. These findings raise the possibility that FGF21 can be used to extend lifespan in other species.DOI:http://dx.doi.org/10.7554/eLife.00065.001.
Cellular granules lacking boundary membranes harbor RNAs and their associated proteins and play diverse roles controlling the timing and location of protein synthesis. Formation of such granules was ...emulated by treatment of mouse brain extracts and human cell lysates with a biotinylated isoxazole (b-isox) chemical. Deep sequencing of the associated RNAs revealed an enrichment for mRNAs known to be recruited to neuronal granules used for dendritic transport and localized translation at synapses. Precipitated mRNAs contain extended 3′ UTR sequences and an enrichment in binding sites for known granule-associated proteins. Hydrogels composed of the low complexity (LC) sequence domain of FUS recruited and retained the same mRNAs as were selectively precipitated by the b-isox chemical. Phosphorylation of the LC domain of FUS prevented hydrogel retention, offering a conceptual means of dynamic, signal-dependent control of RNA granule assembly.
Display omitted
Display omitted
► A crystallized small molecule recruits mRNAs into granule-like aggregates ► The granule-like aggregates contain known granule RNAs ► Aggregates from brain lysates are enriched in mRNAs encoding synaptic proteins ► 3′ UTR lengths of recruited RNAs are significantly longer than average
Aggregation of RNA-binding proteins, including FUS, from mammalian cells within hydrogels retains mRNAs previously associated with RNA granules. Phosphorylation of FUS influences mRNA retention suggesting a mechanism for regulating granule assembly and composition.
Accurate diagnosis and prognosis are essential in lung cancer treatment selection and planning. With the rapid advance of medical imaging technology, whole slide imaging (WSI) in pathology is ...becoming a routine clinical procedure. An interplay of needs and challenges exists for computer-aided diagnosis based on accurate and efficient analysis of pathology images. Recently, artificial intelligence, especially deep learning, has shown great potential in pathology image analysis tasks such as tumor region identification, prognosis prediction, tumor microenvironment characterization, and metastasis detection.
In this review, we aim to provide an overview of current and potential applications for AI methods in pathology image analysis, with an emphasis on lung cancer.
We outlined the current challenges and opportunities in lung cancer pathology image analysis, discussed the recent deep learning developments that could potentially impact digital pathology in lung cancer, and summarized the existing applications of deep learning algorithms in lung cancer diagnosis and prognosis.
With the advance of technology, digital pathology could have great potential impacts in lung cancer patient care. We point out some promising future directions for lung cancer pathology image analysis, including multi-task learning, transfer learning, and model interpretation.