[Identification of breast cancer subtypes based on graph convolutional network].

Identification of molecular subtypes of malignant tumors plays a vital role in individualized diagnosis, personalized treatment, and prognosis prediction of cancer patients. The continuous improvement of comprehensive tumor genomics database and the ongoing breakthroughs in deep learning technology have driven further advancements in computer-aided tumor classification. Although the existing classification methods based on gene expression omnibus database take the complexity of cancer molecular classification into account, they ignore the internal correlation and synergism of genes. To solve this problem, we propose a multi-layer graph convolutional network model for breast cancer subtype classification combined with hierarchical attention network. This model constructs the graph embedding datasets of patients' genes, and develops a new end-to-end multi-classification model, which can effectively recognize molecular subtypes of breast cancer. A large number of test data prove the good performance of this new model in the classification of breast cancer subtypes. Compared to the original graph convolutional neural networks and two mainstream graph neural network classification algorithms, the new model has remarkable advantages. The accuracy, weight-F1-score, weight-recall, and weight-precision of our model in seven-category classification has reached 0.851 7, 0.823 5, 0.851 7 and 0.793 6 respectively. In the four-category classification, the results are 0.928 5, 0.894 9, 0.928 5 and 0.865 0 respectively. In addition, compared with the latest breast cancer subtype classification algorithms, the method proposed in this paper also achieved the highest classification accuracy. In summary, the model proposed in this paper may serve as an auxiliary diagnostic technology, providing a reliable option for precise classification of breast cancer subtypes in the future and laying the theoretical foundation for computer-aided tumor classification.

A Novel Peptide COX52-69 Inhibits High Glucose-induced Insulin Secretion by Modulating BK Channel Activity.

BACKGROUND: Excessive insulin is the leading cause of metabolic syndromes besides hyperinsulinemia. Insulin-lowering therapeutic peptides have been poorly studied and warrant urgent attention. OBJECTIVE: The main purpose of this study, was to introduce a novel peptide COX52-69 that was initially isolated from the porcine small intestine and possessed the ability to inhibit insulin secretion under high-glucose conditions by modulating large conductance Ca2+-activated K+ channels (BK channels) activity. METHODS AND RESULTS: Enzyme-linked immunosorbent assay results indicate that COX52-69 supressed insulin release induced by high glucose levels in pancreatic islets and animal models. Furthermore, electrophysiological data demonstrated that COX52-69 can increase BK channel currents and hyperpolarize cell membranes. Thus, cell excitability decreased, corresponding to a reduction in insulin secretion. CONCLUSION: Our study provides a novel approach to modulate high glucose-stimulated insulin secretion in patients with hyperinsulinemia.

Pharmacokinetics of Novel Oral Cyclic Peptide.

Protein and peptide drugs have been considered to be valuable for treating disease for many years, capturing more and more of the attention of researchers. Previously, we found a short peptide from the porcine intestine named COX52-69, which could simultaneously lower blood glucose and insulin response after intraperitoneal injection. And thus, it showed a potential to counter type II diabetes without leading to insulin resistance, mainly caused by high insulin levels in the blood. However, this molecule is not stable in the digestive system and cannot be used via oral administration. Here we employed the circularization technique to modify the peptide and tested its pharmacokinetics.

Amygdala Metabotropic Glutamate Receptor 1 Influences Synaptic Transmission to Participate in Fentanyl-Induced Hyperalgesia in Rats.

The underlying mechanisms of opioid-induced hyperalgesia (OIH) remain unclear. Herein, we found that the protein expression of metabotropic glutamate receptor 1 (mGluR1) was significantly increased in the right but not in the left laterocapsular division of central nucleus of the amygdala (CeLC) in OIH rats. In CeLC neurons, the frequency and the amplitude of mini-excitatory postsynaptic currents (mEPSCs) were significantly increased in fentanyl group which were decreased by acute application of a mGluR1 antagonist, A841720. Finally, the behavioral hypersensitivity could be reversed by A841720 microinjection into the right CeLC. These results show that the right CeLC mGluR1 is an important factor associated with OIH that enhances synaptic transmission and could be a potential drug target to alleviate fentanyl-induced hyperalgesia.

ACP-DA: Improving the Prediction of Anticancer Peptides Using Data Augmentation.

Anticancer peptides (ACPs) have provided a promising perspective for cancer treatment, and the prediction of ACPs is very important for the discovery of new cancer treatment drugs. It is time consuming and expensive to use experimental methods to identify ACPs, so computational methods for ACP identification are urgently needed. There have been many effective computational methods, especially machine learning-based methods, proposed for such predictions. Most of the current machine learning methods try to find suitable features or design effective feature learning techniques to accurately represent ACPs. However, the performance of these methods can be further improved for cases with insufficient numbers of samples. In this article, we propose an ACP prediction model called ACP-DA (Data Augmentation), which uses data augmentation for insufficient samples to improve the prediction performance. In our method, to better exploit the information of peptide sequences, peptide sequences are represented by integrating binary profile features and AAindex features, and then the samples in the training set are augmented in the feature space. After data augmentation, the samples are used to train the machine learning model, which is used to predict ACPs. The performance of ACP-DA exceeds that of existing methods, and ACP-DA achieves better performance in the prediction of ACPs compared with a method without data augmentation. The proposed method is available at http://github.com/chenxgscuec/ACPDA.

Isoforms of the nuclear envelope protein Nurim are differentially expressed during heart development in mice.

To date, transcript variants of the nuclear envelope protein Nurim and their expression profiles in mice have never been elucidated. In this study, we determined that the primary Nurim variant a was abundantly expressed in mouse heart, liver, spleen and kidney. The protein level of isoform a is initiated at an early stage of heart formation and demonstrated a significant increase in expression throughout embryonic heart development. Interestingly, Nurim b is also up-regulated from E12.5 to E18.5 in different individuals. Our research represents the first report on alternative splicing variants of mouse Nurim and their differential expression profile during embryonic development. These studies suggest a potential role for Nurim in early heart morphogenesis and should help further elucidate the function of Nurim.

Integrative analysis of gene expression associated with epilepsy in human epilepsy and animal models.

Epilepsy is a severe neuropsychiatric disorder, the cause of which remains to be elucidated. Genome­wide association studies, DNA microarrays and proteomes have been widely applied to identify the candidate genes involved in epileptogenesis, and integrative analyses are often capable of extracting more detailed biological information from the data. In the present study, a total number of 1,065 genes in different animal models were collected to construct an epilepsy candidate gene database. Further analyses suggested that the response to organic substances, the intracellular signaling cascade and neurological system processes were significantly enriched biological processes, and the mitogen-activated protein kinase pathway was identified as a putative epileptogenic signaling pathway. In addition, the five key genes, growth factor receptor bound 2, amyloid ß (A4) precursor protein, transforming growth factor­ß, vascular endothelial growth factor and cyclin­dependent kinase inhibitor 1, were identified as being critical as central nodes in the protein networks. Reverse transcription­quantitative polymerase chain reaction analysis revealed that these genes were all upregulated at the mRNA level in the epileptic loci compared with the resection margin of tissue samples from the same patients diagnosed with epilepsy. The data mining performed in the present study thus was shown to be a useful tool, which may contribute to obtaining further information on epileptic disorders and delineating the molecular mechanism of the associated genes.

Characteristics of hyperpolarization-activated cyclic nucleotide-gated channels in dorsal root ganglion neurons at different ages and sizes.

In rat's sensory neurons, hyperpolarization-activated inward currents (Ih) play an essential role in mediating action potentials and contributing to neuronal excitability. Classified by the size of neurons and ages, we studied the Ih and transcription levels of hyperpolarization-activated cyclic nucleotide-gated (HCN) channels using electrophysiology and the single-cell RT-PCR. In voltage-clamp studies, Ih and half-maximal activation voltage (V1/2) changed with age and size. An analysis of all HCN subtypes in dorsal root ganglion (DRG) neurons by single-cell RT-PCR was carried out. HCN1 and HCN3 in medium-small elderly neurons had a weak expression. HCN2 in newborns and HCN4 in elderly rats also had a weak expression. The aim of this study is to examine the age-related Ih and HCN channels subunits in different ages and sizes of DRG neurons. The results would be significant in understanding the physiological and pathophysiological function of different sizes of DRG neurons in different age periods.

A novel testis-enriched gene Spata33 is expressed during spermatogenesis.

With an increasing incidence of male idiopathic infertility, identification of novel genes involved in spermatogenesis is an important aspect for the understanding of human testicular failure. In the present study, we have identified a novel gene Spata33, also called as 4732415M23Rik or C16orf55, which is conserved in mammalian species. Spata33 was predominantly expressed in the postpartum and adult mouse testes at mRNA and protein levels. Its expression was increased during the first wave of the spermatogenesis, indicating that Spata33 may be associated with the meiotic process. Further immunohistochemistry analysis revealed that Spata33 was mainly expressed in the spermatocytes, spermatogonia and round spermatids. Its expression was uniformly distributed in the nucleus and cytosol in these germ cells, which was further confirmed by Spata33-tagged with GFP staining in the GC-1 and TM4 cells. These results indicated that Spata33 was predominantly expressed in the mouse testis and associated with spermatogenesis. Identification and characterization of the novel testis-enriched gene Spata33 may provide a new route for understanding of spermatogenesis failure.

Genome-wide detection of testis- and testicular cancer-specific alternative splicing.

Alternative pre-messenger RNA (mRNA) splicing is a key molecular event that allows for protein diversity and plays important roles in development and disease. Alternative pre-mRNA splicing regulations during spermatogenesis and alternative pre-mRNA splicing etiology in testicular tumorigenesis are yet to be characterized. By genome-wide analysis, here we describe alternative splicing features that distinguish distinctive patterns of alternative pre-mRNA splicing among human testis, testicular cancer and mouse testis. Through computationally subtractive analysis, we detected 80 testis-specific transcript candidates in human testis, 175 in human testicular cancer and 262 in mouse testis, which were integrated into a database. Reverse transcription-polymerase chain reaction confirmed that most of these transcript candidates from mouse testis were testis specific. Around 40% of the transcripts were from unknown/hypothetical genes, which were useful for further functional analysis. These transcripts were not overlapped, indicating lack of evolutionary conservation. Further chromosome mapping showed distinct chromosomal preference of alternative pre-mRNA splicing events. Comparison analysis indicated that alternative pre-mRNA splicing in human testicular tumor shared some characters/trends with those in mouse testis. Moreover, human testicular tumor tended to use rare splice sites and there were also distinct sequences adjacent dominant splice sites between normal testis and testicular tumor. These special features of alternative pre-mRNA splicing in human testicular tumor suggested that testicular tumorigenesis was involved in multiple steps/levels of alternative splicing events. Using alternative splicing as a potential source for new clinical diagnostic, prognostic and therapeutic strategies for treatment of testicular tumors seems to have a bright prospect.