HSG-MGAF Net: Heterogeneous subgraph-guided multiscale graph attention fusion network for interpretable prediction of whole-slide image.

BACKGROUND AND OBJECTIVE: Pathological whole slide image (WSI) prediction and region of interest (ROI) localization are important issues in computer-aided diagnosis and postoperative analysis in clinical applications. Existing computer-aided methods for predicting WSI are mainly based on multiple instance learning (MIL) and its variants. However, most of the methods are based on instance independence and identical distribution assumption and performed at a single scale, which not fully exploit the hierarchical multiscale heterogeneous information contained in WSI. METHODS: Heterogeneous Subgraph-Guided Multiscale Graph Attention Fusion Network (HSG-MGAF Net) is proposed to build the topology of critical image patches at two scales for adaptive WSI prediction and lesion localization. The HSG-MGAF Net simulates the hierarchical heterogeneous information of WSI through graph and hypergraph at two scales, respectively. This framework not only fully exploits the low-order and potential high-order correlations of image patches at each scale, but also leverages the heterogeneous information of the two scales for adaptive WSI prediction. RESULTS: We validate the superiority of the proposed method on the CAMELYON16 and the TCGA- NSCLC, and the results show that HSG-MGAF Net outperforms the state-of-the-art method on both datasets. The average ACC, AUC and F1 score of HSG-MGAF Net can reach 92.7 %/0.951/0.892 and 92.2 %/0.957/0.919, respectively. The obtained heatmaps can also localize the positive regions more accurately, which have great consistency with the pixel-level labels. CONCLUSIONS: The results demonstrate that HSG-MGAF Net outperforms existing weakly supervised learning methods by introducing critical heterogeneous information between the two scales. This approach paves the way for further research on light weighted heterogeneous graph-based WSI prediction and ROI localization.

Interpretable Inference and Classification of Tissue Types in Histological Colorectal Cancer Slides Based on Ensembles Adaptive Boosting Prototype Tree.

Digital pathology images are treated as the "gold standard" for the diagnosis of colorectal lesions, especially colon cancer. Real-time, objective and accurate inspection results will assist clinicians to choose symptomatic treatment in a timely manner, which is of great significance in clinical medicine. However, Manual methods suffers from long inspection cycle and serious reliance on subjective interpretation. It is also a challenging task for existing computer-aided diagnosis methods to obtain models that are both accurate and interpretable. Models that exhibit high accuracy are always more complex and opaque, while interpretable models may lack the necessary accuracy. Therefore, the framework of ensemble adaptive boosting prototype tree is proposed to predict the colorectal pathology images and provide interpretable inference by visualizing the decision-making process in each base learner. The results showed that the proposed method could effectively address the "accuracy-interpretability trade-off" issue by ensemble of m adaptive boosting neural prototype trees. The superior performance of the framework provides a novel paradigm for interpretable inference and high-precision prediction of pathology image patches in computational pathology.

Effects of different concentrations of androgens on KAP24.1 gene expression in Hetian sheep and Karakul sheep.

The purpose of the experiment was to clone and eukaryotic expression of hair follicle keratin associated protein 24.1 (KAP24.1), study the effect of different concentrations of androgen on protein expression, and compare KAP24.1 gene in skin and hair follicles of different breeds of sheep expression, explore KAP24.1 Expression difference of gene among local sheep breeds in southern Xinjiang and its effect on wool quality. The body-side hair follicles of Plain-type Hetian sheep, Mountain-type Hetian sheep and Karakul sheep were used as experimental materials, and the KAP24.1 gene sequence of sheep in GenBank (accession number: JX112014.1) was used as the reference to design primers. The KAP24.1 gene was amplified by PCR, and the pMD19-T-KAP24.1 cloning plasmid was constructed. After double digestion and identification, the pEGFP-N1-KAP24.1 eukaryotic recombinant expression plasmid was constructed. After PCR and double digestion and identification, sequencing and sequence analysis were performed, and the expression was transfected into Hela cells. SDS-PAGE and Western blotting were used to detect the expression levels of androgen at different concentrations. The expression of KAP24.1 gene in different sheep skin follicles was detected by real-time fluorescent quantitative PCR. Three sheep KAP24.1 were cloned The CDS region sequence of gene is 759 bp, encoding 252 amino acids, all of which are unstable hydrophobic proteins.The results of similarity comparison showed that compared with the reference gene, the gene sequence similarity of Mountain-type Hetian sheep and Karakul Sheep was 99.47%, and that of Plain-type Hetian sheep was 99.34%. Phylogenetic tree analysis showed that the three sheep had the closest genetic relationship with Capra hircus and the furthest genetic relationship with Cervus canadensis.The secondary structure of KAP24.1 was mainly composed of random coil.PEGFP-N1-KAP24.1 was successfully constructed eukaryotic recombinant expression plasmid was successfully transfected into HeLa cells to obtain 58 kDa KAP24.1 recombinant protein. When the concentration of androgen is 10-8 mol / L, the protein expression is the highest. The expression of KAP24.1 gene in skin and hair follicles of Mountain-type Hetian sheep was significantly different from that of plain-type Hetian sheep (P < 0.05), and there was significant difference between Mountain-type Hetian sheep and Karakul Sheep (P < 0.05). The expression of Karakul Sheep was significantly higher than that of Plain-type Hetian sheep (P < 0.05). The 759 bp CDS sequence of KAP24.1 gene in sheep was cloned, and PEGFP-N1-KAP24.1 was constructed eukaryotic recombinant expression plasmid to obtain 58 kDa KAP24.1 recombinant protein. When the concentration of androgen was 10-8mol / L, the protein expression was the highest, and KAP24.1 gene was expressed in the skin and hair follicles of three sheep breeds, and the expression of Mountain-type Hetian sheep was the highest.

Interpretable classification of pathology whole-slide images using attention based context-aware graph convolutional neural network.

BACKGROUND AND OBJECTIVE: Whole slide image (WSI) classification and lesion localization within giga-pixel slide are challenging tasks in computational pathology that requires context-aware representations of histological features to adequately infer nidus. The existing weakly supervised learning methods mainly treat different locations in the slide as independent regions and cannot learn potential nonlinear interactions between instances based on i.i.d assumption, resulting in the model unable to effectively utilize context-ware information to predict the labels of WSIs and locate the region of interest (ROI). METHODS: Here, we propose an interpretable classification model named bidirectional Attention-based Multiple Instance Learning Graph Convolutional Network (ABMIL-GCN), which hierarchically aggregates context-aware features of instances into a global representation in a topology fashion to predict the slide labels and localize the region of lymph node metastasis in WSIs. RESULTS: We verified the superiority of this method on the Camelyon16 dataset, and the results show that the average predicted ACC and AUC of the proposed model after flooding optimization can reach 90.89% and 0.9149, respectively. The average accuracy and ACC score are improved by more than 7% and 4% compared with the existing state-of-the-art algorithms. CONCLUSIONS: The results demonstrate that context-aware GCN outperforms existing weakly supervised learning methods by introducing spatial correlations between the neighbor image patches, which also addresses the 'accuracy-interpretability trade-off' problem. The framework provides a novel paradigm for the clinical application of computer-aided diagnosis and intelligent systems.

A novel strategy regarding geometric product for liquids discrimination based on THz reflection spectroscopy.

A novel expression of geometric product that associated with the geometric relationship from geometric algebra constructed by the vectorized refractive index and absorption coefficient in THz region is proposed, which could provide a new insight into the THz properties of materials. From the novel expression, the candidate characteristic parameters are extracted for liquids discrimination and present the abundant second order correlation information of optical parameters with the consideration of dimension rising. Three groups of liquids, containing C-reactive protein calibrators and alpha fetoprotein calibrators, were selected as examples to validate the feasibility of the proposed strategy. Comparing with the traditional THz parameters, including refractive index, absorption coefficient, and complex permittivity, the novel approach exhibits notable superiority for differentiation with the evaluation of statistical differences and effect sizes. The proposed geometric product expression could have a large potential on promoting the substance identification in some applications of THz technology.

The chromatin accessibility landscape reveals distinct transcriptional regulation in the induction of human primordial germ cell-like cells from pluripotent stem cells.

In vitro induction of human primordial germ cell-like cells (hPGCLCs) provides an ideal platform to recapitulate hPGC development. However, the detailed molecular mechanisms regulating the induction of hPGCLCs remain largely uncharacterized. Here, we profiled the chromatin accessibility and transcriptome dynamics throughout the process of hPGCLC induction. Genetic ablation of SOX15 indicated the crucial roles of SOX15 in the maintenance of hPGCLCs. Mechanistically, SOX15 exerted its roles via suppressing somatic gene expression and sustaining latent pluripotency. Notably, ETV5, a downstream regulator of SOX15, was also uncovered to be essential for hPGCLC maintenance. Finally, a stepwise switch of OCT4/SOX2, OCT4/SOX17, and OCT4/SOX15 binding motifs were found to be enriched in closed-to-open regions of human embryonic stem cells, and early- and late-stage hPGCLCs, respectively. Collectively, our data characterized the chromatin accessibility and transcriptome landscapes throughout hPGCLC induction and defined the SOX15-mediated regulatory networks underlying this process.

Diagnosis of hepatocellular carcinoma based on a terahertz signal and VMD-CWSE.

A novel strategy on combining variational mode decomposition (VMD) and composite weighted-scale sample entropy (CWSE) modified from composite multiscale entropy (CMSE) is proposed to screen hepatocellular carcinoma (HCC) by measuring the terahertz (THz) pulse signals of ten normal and ten HCC serums. Eight measured HCC specimens are negative in serum biomarker alpha fetoprotein (AFP) determination. In CWSE, the time series with weighted-scales are generated from the weighted average processing in the coarse-grained time series corresponding to each scale of the CMSE algorithm. VMD served as a preprocessing method was introduced into decomposing THz signal to obtain the mode functions of specific bandwidth for identification. Final results reveal that more obtainable entropy values of CWSE for recognition in comparison to those of CMSE on the basis of the rule of statistically significant difference and effect size and also manifest the stronger discriminability than the traditional THz parameters. This study provides a new potential auxiliary tool for diagnosis HCC and develops the methodology on the discrimination for similar THz signals.

Analysis of the Prognosis and Therapeutic Value of the CXC Chemokine Family in Head and Neck Squamous Cell Carcinoma.

The CXC chemokines belong to a family which includes 17 different CXC members. Accumulating evidence suggests that CXC chemokines regulate tumor cell proliferation, invasion, and metastasis in various types of cancers by influencing the tumor microenvironment. The different expression profiles and specific function of each CXC chemokine in head and neck squamous cell carcinoma (HNSCC) are not yet clarified. In our work, we analyzed the altered expression, interaction network, and clinical data of CXC chemokines in patients with HNSCC by using the following: the Oncomine dataset, cBioPortal, Metascape, String analysis, GEPIA, and the Kaplan-Meier plotter. The transcriptional level analysis suggested that the mRNA levels of CXCL1, CXCL2, CXCL3, CXCL5, CXCL6, CXCL8, CXCL9, CXCL10, CXCL11, and CXCL13 increased in HNSCC tissue samples when compared to the control tissue samples. The expression levels of CXCL9, CXCL10, CXCL11, CXCL12, and CXCL14 were associated with various tumor stages in HNSCC. Clinical data analysis showed that high transcription levels of CXCL2, CXCL3, and CXCL12, were linked with low relapse-free survival (RFS) in HNSCC patients. On the other hand, high CXCL14 levels predicted high RFS outcomes in HNSCC patients. Meanwhile, increased gene transcription levels of CXCL9, CXCL10, CXCL13, CXCL14, and CXCL17 were associated with a higher overall survival (OS) advantage in HNSCC patients, while high levels of CXCL1, and CXCL8 were associated with poor OS in all HNSCC patients. This study implied that CXCL1, CXCL2, CXCL3, CXCL8, and CXCL12 could be used as prognosis markers to identify low survival rate subgroups of patients with HNSCC as well as be potential suitable therapeutic targets for HNSCC patients. Additionally, CXCL9, CXCL10, CXCL13, CXCL14, and CXCL17 could be used as functional prognosis biomarkers to identify better survival rate subgroups of patients with HNSCC.

[Serum galectin-9 in cervical cancer].

OBJECTIVE: To investigate the expression level of Galectin-9 (Gal-9) in the serum of cervical cancer patients and analyze the clinic significance of it. METHODS: Collected sera from 29 patients with cervical cancer, 45 patients with cervical inter-epithelial lesion (CIN) and 26 normal females. Determination of Gal-9 was performed by ELISA, and in the group of patients with cervical cancer, these results were compared with the serum vascular endothelial growth factor (VEGF) and C-reactive protein (CRP). Then selecting four patients from the invasive cervical cancer group performed Gal-9 retrospective immuno-staining analysis. RESULTS: Gal-9 concentration was 27.15 ng/L in the group of patients with cervical cancer that was significantly higher compared with the patients with CIN (21.20 ng/L) and the normal females (17.43 ng/L), and the Gal-9 concentration was significantly correlated with both VEGF and CRP in cervical cancer group. Two of these patients, who had a higher Gal-9 serum level, shown more intense staining and the other two patients, with lower serum level of Gal-9, had moderate immunostaining, suggesting that at least part of serum Gal-9 might be produced by cervical cancer tissue. CONCLUSIONS: Gal-9 might play a role in the progression and invasion of cervical cancer and warrants further study.

Galectin-9 expression links to malignant potential of cervical squamous cell carcinoma.

PURPOSE: Galectin-9 (Gal-9) induces adhesion and aggregation of certain cell types and can be a prognostic factor in the patients with melanoma and breast cancer. We assessed the experiments to resolve whether Gal-9 expression in cervical neoplasm links to malignant potential of cervical squamous cell carcinoma (SCC) cells. METHODS: Gal-9 expression was examined with immunohistochemical techniques in 23 normal cervical squamous epithelia, 17 cervical intraepithelial neoplasia (CIN), and 38 cervical SCC compared to E-cadherin. CIN was divided into low-grade and high-grade squamous intraepithelial lesions (8 LSIL and 9 HSIL), and SCC was into well-, moderately and poorly differentiated SCC (6 WSCC, 20 MSCC and 12 PSCC). RESULTS: Gal-9 and E-cadherin were evidently detected in normal epithelium and endocervical glands, but those in CIN and SCC were significantly faint. Moreover, both the Gal-9 and E-cadherin expressions in HSIL were significant lower than those in LSIL, suggesting their association with malignant transformation. Unexpectedly, Gal-9 and E-cadherin in WSCC were significantly high compared to those in HSIL. Furthermore, those in SCC were inversely correlated with the grade of differentiation (WSCC >> MSCC >> PSCC), implying the possible involvement of Gal-9 and E-cadherin in the differentiation of SCC. In contrast, they were not different among the FIGO stage. Gal-9 expression was well correlated with E-cadherin expression in CIN and SCC but not in normal cervical epithelia. CONCLUSION: The present results suggest that decreased Gal-9 expression is inversely associated with malignant potential or differentiation of cervical CIN and SCC as a differentiation biomarker.

[SiRNA inhibition of E6AP expression in cervical cancer cells].

OBJECTIVE: To study the effect of RNA interference (RNAi) targeting E6AP on the proliferation and apoptosis of HeLa cells. METHODS: HeLa cells were cultured and divided into 3 groups: blank control group, cells transfected with nonsense siRNA (small interference RNA), and cells transfected with specific E6AP siRNA. The expressions of E6AP mRNA and protein were detected by RT-PCR and Western blot before and after the transfection respectively. Cell proliferation was determined by methylthiazolyl tetrazolium (MTT). The cell apoptosis index was assessed by flow cytometry. RESULTS: Upon treatment with E6AP siRNA for 24, 48 and 72 h, the expression level of E6AP mRNA decreased 33%, 72% and 70% than siRNA treated group. The protein expression levels in 48 h and 72 h E6AP siRNA groups decreased 38%, 59% comparing with those of the nonsense siRNA treated group (P < 0.05). The proliferative capacity of cells transfectd with E6AP siRNA was significantly lower than blank control group (F = 101.38, P < 0.05) and siRNA treated group (F = 38.64, P < 0.05). The apoptosis index of HeLa cells treated with E6AP siRNA was significantly higher than that of the nonsense siRNA (F = 41.48, P < 0.05) and the blank control group (F = 86.36, P < 0.05). CONCLUSION: SiRNA targeting can effectively suppress the expression levels of E6AP mRNA, corresponding with a proliferation inhibition and an enhanced apoptosis of HeLa cells.