Asthma prediction via affinity graph enhanced classifier: a machine learning approach based on routine blood biomarkers.

BACKGROUND: Asthma is a chronic respiratory disease affecting millions of people worldwide, but early detection can be challenging due to the time-consuming nature of the traditional technique. Machine learning has shown great potential in the prompt prediction of asthma. However, because of the inherent complexity of asthma-related patterns, current models often fail to capture the correlation between data samples, limiting their accuracy. Our objective was to use our novel model to address the above problem via an Affinity Graph Enhanced Classifier (AGEC) to improve predictive accuracy. METHODS: The clinical dataset used in this study consisted of 152 samples, where 24 routine blood markers were extracted as features to participate in the classification due to their ease of sourcing and relevance to asthma. Specifically, our model begins by constructing a projection matrix to reduce the dimensionality of the feature space while preserving the most discriminative features. Simultaneously, an affinity graph is learned through the resulting subspace to capture the internal relationship between samples better. Leveraging domain knowledge from the affinity graph, a new classifier (AGEC) is introduced for asthma prediction. AGEC's performance was compared with five state-of-the-art predictive models. RESULTS: Experimental findings reveal the superior predictive capabilities of AGEC in asthma prediction. AGEC achieved an accuracy of 72.50%, surpassing FWAdaBoost (61.02%), MLFE (60.98%), SVR (64.01%), SVM (69.80%) and ERM (68.40%). These results provide evidence that capturing the correlation between samples can enhance the accuracy of asthma prediction. Moreover, the obtained [Formula: see text] values also suggest that the differences between our model and other models are statistically significant, and the effect of our model does not exist by chance. CONCLUSION: As observed from the experimental results, advanced statistical machine learning approaches such as AGEC can enable accurate diagnosis of asthma. This finding holds promising implications for improving asthma management.

Extraordinarily Time-and Memory-Efficient Large-Scale Canonical Correlation Analysis in Fourier Domain: From Shallow to Deep.

Canonical correlation analysis (CCA) is a correlation analysis technique that is widely used in statistics and the machine-learning community. However, the high complexity involved in the training process lays a heavy burden on the processing units and memory system, making CCA nearly impractical in large-scale data. To overcome this issue, a novel CCA method that tries to carry out analysis on the dataset in the Fourier domain is developed in this article. Appling Fourier transform on the data, we can convert the traditional eigenvector computation of CCA into finding some predefined discriminative Fourier bases that can be learned with only element-wise dot product and sum operations, without complex time-consuming calculations. As the eigenvalues come from the sum of individual sample products, they can be estimated in parallel. Besides, thanks to the data characteristic of pattern repeatability, the eigenvalues can be well estimated with partial samples. Accordingly, a progressive estimate scheme is proposed, in which the eigenvalues are estimated through feeding data batch by batch until the eigenvalues sequence is stable in order. As a result, the proposed method shows its characteristics of extraordinarily fast and memory efficiencies. Furthermore, we extend this idea to the nonlinear kernel and deep models and obtained satisfactory accuracy and extremely fast training time consumption as expected. An extensive discussion on the fast Fourier transform (FFT)-CCA is made in terms of time and memory efficiencies. Experimental results on several large-scale correlation datasets, such as MNIST8M, X-RAY MICROBEAM SPEECH, and Twitter Users Data, demonstrate the superiority of the proposed algorithm over state-of-the-art (SOTA) large-scale CCA methods, as our proposed method achieves almost same accuracy with the training time of our proposed method being 1000 times faster. This makes our proposed models best practice models for dealing with large-scale correlation datasets. The source code is available at https://github.com/Mrxuzhao/FFTCCA.

The role of calcineurin/NFAT in SFRP2 induced angiogenesis--a rationale for breast cancer treatment with the calcineurin inhibitor tacrolimus.

Tacrolimus (FK506) is an immunosuppressive drug that binds to the immunophilin FKBPB12. The FK506-FKBP12 complex associates with calcineurin and inhibits its phosphatase activity, resulting in inhibition of nuclear translocation of nuclear factor of activated T-cells (NFAT). There is increasing data supporting a critical role of NFAT in mediating angiogenic responses stimulated by both vascular endothelial growth factor (VEGF) and a novel angiogenesis factor, secreted frizzled-related protein 2 (SFRP2). Since both VEGF and SFRP2 are expressed in breast carcinomas, we hypothesized that tacrolimus would inhibit breast carcinoma growth. Using IHC (IHC) with antibodies to FKBP12 on breast carcinomas we found that FKBP12 localizes to breast tumor vasculature. Treatment of MMTV-neu transgenic mice with tacrolimus (3 mg/kg i.p. daily) (n = 19) resulted in a 73% reduction in the growth rate for tacrolimus treated mice compared to control (n = 15), p = 0.003; which was associated with an 82% reduction in tumor microvascular density (p<0.001) by IHC. Tacrolimus (1 µM) inhibited SFRP2 induced endothelial tube formation by 71% (p = 0.005) and inhibited VEGF induced endothelial tube formation by 67% (p = 0.004). To show that NFATc3 is required for SFRP2 stimulated angiogenesis, NFATc3 was silenced with shRNA in endothelial cells. Sham transfected cells responded to SFRP2 stimulation in a tube formation assay with an increase in the number of branch points (p<0.003), however, cells transfected with shRNA to NFATc3 showed no increase in tube formation in response to SFRP2. This demonstrates that NFATc3 is required for SFRP2 induced tube formation, and tacrolimus inhibits angiogenesis in vitro and breast carcinoma growth in vivo. This provides a rationale for examining the therapeutic potential of tacrolimus at inhibiting breast carcinoma growth in humans.

Molecular characterization of mucosal adherent bacteria and associations with colorectal adenomas.

The human large bowel is colonized by complex and diverse bacterial communities. However, the relationship between commensal bowel bacteria and adenomas (colorectal cancer precursors) is unclear. This study aimed to characterize adherent bacteria in normal colon and evaluate differences in community composition associated with colorectal adenomas. We evaluated adherent bacteria in normal colonic mucosa of 21 adenoma and 23 non-adenoma subjects enrolled in a cross sectional study. Terminal restriction fragment length polymorphism, clone sequencing and fluorescent in-situ hybridization analysis of the 16S rRNA genes were used to characterize adherent bacteria. A total of 335 clones were sequenced and processed for phylogenetic and taxonomic analysis. Differences in bacterial composition between cases and controls were evaluated by UniFrac and analysis of similarity matrix. Overall, Firmicutes (62%), Bacteroidetes (26%) and Proteobacteria (11%) were the most dominant phyla. The bacterial composition differed significantly between cases and controls (UniFrac p < 0.001). We observed significantly higher abundance of Proteobacteria (p < 0.05) and lower abundance of Bacteroidetes (p < 0.05) in cases compared to controls. At the genus level, case subjects showed increased abundance of Dorea spp. (p < 0.005), Faecalibacterium spp. (p < 0.05) and lower proportions of Bacteroides spp. (p < 0.03) and Coprococcus spp. (p < 0.05) than controls. Cases had higher bacterial diversity and richness than controls. These findings reveal that alterations in bacterial community composition associated with adenomas may contribute to the etiology of colorectal cancer. Extension of these findings could lead to strategies to manipulate the microbiota to prevent colorectal adenomas and cancer as well as to identify individuals at high risk.

beta3-adrenergic receptor induction of adipocyte inflammation requires lipolytic activation of stress kinases p38 and JNK.

Activation of beta-adrenergic receptors (AR) in adipocytes triggers acute changes in metabolism that can alter patterns of gene expression. This work examined the mechanisms by which activation of hormone sensitive lipase (HSL) induces expression of inflammatory cytokines in adipocytes in vivo and model adipocytes in vitro. beta3-AR activation in mice triggered expression of inflammatory genes CCL2, IL-6, and PAI-1, as well as endoplasmic reticulum (ER) stress markers GRP78 and CHOP. Pharmacological inhibition of HSL blocked induction of inflammatory genes, but not ER stress markers. Promoting intracellular accumulation of free fatty acids (FFAs) in 3T3-L1 adipocytes increased expression of inflammatory cytokines, whereas inhibiting ceramide synthesis partly blocked PAI-1 expression, but not IL-6. Induction of inflammatory markers in vivo and in vitro was preceded by phosphorylation of p38 and JNK, and inhibition of HSL prevented activation of these kinases. Experiments with pharmacological inhibitors of specific MAP kinases demonstrated the importance of p38 MAPK as a mediator of lipolysis-induced inflammation in vivo and in vitro. Together, these results demonstrate that FFAs liberated by HSL activate p38 and JNK, and p38 mediates pro-inflammatory cytokine expression in adipose tissue.

KRAS/BRAF mutation status and ERK1/2 activation as biomarkers for MEK1/2 inhibitor therapy in colorectal cancer.

Phase II clinical trials of mitogen-activated protein/extracellular signal-regulated kinase (ERK) kinase (MEK) inhibitors are ongoing and ERK1/2 activation is frequently used as a biomarker. In light of the mutational activation of BRAF and KRAS in colorectal cancer, inhibitors of the Raf-MEK-ERK mitogen-activated protein kinase are anticipated to be promising. Previous studies in pancreatic cancer have found little correlation between BRAF/KRAS mutation status and ERK1/2 activation, suggesting that identifying biomarkers of MEK inhibitor response may be more challenging than previously thought. The purpose of this study was to evaluate the effectiveness of MEK inhibitor therapy for colorectal cancer and BRAF/KRAS mutation status and ERK1/2 activation as biomarkers for MEK inhibitor therapy. First, we found that MEK inhibitor treatment impaired the anchorage-independent growth of nearly all KRAS/BRAF mutant, but not wild-type, colorectal cancer cells. There was a correlation between BRAF, but not KRAS, mutation status and ERK1/2 activation. Second, neither elevated ERK1/2 activation nor reduction of ERK1/2 activity correlated with MEK inhibition of anchorage-independent growth. Finally, we validated our cell line observations and found that ERK1/2 activation correlated with BRAF, but not KRAS, mutation status in 190 patient colorectal cancer tissues. Surprisingly, we also found that ERK activation was elevated in normal colonic epithelium, suggesting that normal cell toxicity may be a complication for colorectal cancer treatment. Our results suggest that although MEK inhibitors show promise in colorectal cancer, KRAS/BRAF mutation status, but not ERK activation as previously thought, may be useful biomarkers for MEK inhibitor sensitivity.

Role of hormone-sensitive lipase in beta-adrenergic remodeling of white adipose tissue.

Free fatty acids (FFA) are important extracellular and intracellular signaling molecules and are thought to be involved in beta-adrenergic-induced remodeling of adipose tissue, which involves a transient inflammatory response followed by mitochondrial biogenesis and increased oxidative capacity. This work examined the role of hormone-sensitive lipase (HSL), a key enzyme of acylglycerol metabolism, in white adipose tissue (WAT) remodeling using genetic inactivation or pharmacological inhibition. Acute treatment with the beta(3)-adrenergic agonist CL-316,243 (CL) induced expression of inflammatory markers and caused extravasation of myeloid cells in WAT of wild-type (WT) mice. HSL-knockout (KO) mice had elevated inflammatory gene expression in the absence of stimulation, and acute injection of CL did not further recruit myeloid cells, nor did it further elevate inflammatory gene expression. Acute pharmacological inhibition of HSL with BAY 59-9435 (BAY) had no effect on inflammatory gene expression in WAT or in cultured 3T3-L1 adipocytes. However, BAY prevented induction of inflammatory cytokines by beta-adrenergic stimulation in WAT in vivo and in cultured 3T3-L1 adipocytes. Chronic CL treatment stimulated mitochondrial biogenesis, expanded oxidative capacity, and increased lipid droplet fragmentation in WT mice, and these effects were significantly impaired in HSL-KO mice. In contrast to HSL-KO mice, mice with defective signaling of Toll-like receptor 4, a putative FFA receptor, showed normal beta-adrenergic-induced remodeling of adipose tissue. Overall, results reveal the importance of HSL activity in WAT metabolic plasticity and inflammation.

Phylogenetic analysis in chicken breeds inferred from complete cytochrome b gene information.

Complete cytochrome b (cyt b) gene (1140 bp) nucleotide sequences were used to investigate characteristics of the genetic constitution of Chunky broiler chickens, and these were compared with the Hy-Line and WL-GM (Garber) line of White Leghorn, the GSP line of Fayoumi, the BM-C line of Black Minorca (egg-chickens), and an outgroup of wild-origin Japanese quail. A high genetic difference (five haplotypes) was observed at the cytochrome b region in the Chunky broiler in contrast to the high homologies observed among the other chicken breeds (egg-purpose). Chunky broilers can be distinguished from the other breeds (White Leghorn, Fayoumi, and Black Minorca) at positions 552 and 779. The molecular phylogenetic tree exhibited genetic differences within Chunky broilers, and between Chunky broilers and the other three chicken breeds. As a result, some chicken strains or breeds apparently different from the other egg-chickens may have contributed to the Chunky broiler formation. Artificial selection may be one of the biggest factors causing nucleotide diversity in the chicken breeds.