MOPDDI: Predicting Drug-Drug Interaction Events Based on Multimodal Mutual Orthogonal Projection and Intermodal Consistency Loss.

Predicting accurately the mechanisms of drug-drug interaction (DDI) events is crucial in drug research and development. Existing methods used to predict these events are primarily based on deep learning and have achieved satisfactory results. However, they rarely consider the presence of redundant co-information between the multimodal data of a drug and the need for consistency in the predicted features of each drug modality. Herein, we propose a new method for drug interaction event prediction based on multimodal mutual orthogonal projection and intermodal consistency loss. Our method obtains the features of each modality through a multimodal mutual orthogonal projection module, which eliminates redundant common information with other modalities. In addition, we use the consistency loss between modalities and make the predicted features of each modality more similar. In comparative experiments, our proposed method achieves a prediction accuracy of 0.9500, and an area under the precision-recall (AUPR) curve is 0.9833 for known DDIs. This method outperforms existing methods. The results show that the proposed method is capable of accurately predicting DDIs. The source code is available at https://github.com/xiaqixiaqi/MOPDDI.

Matrine restrains the development of colorectal cancer through regulating the AGRN/Wnt/ß-catenin pathway.

BACKGROUND: Colorectal cancer is a common malignant digestive tract tumor. This study aimed to explore the biological role and potential underlying mechanism of matrine in colorectal cancer. METHODS: The mRNA expression of AGRN was measured using RT-qPCR. Cell proliferation, migration, invasion and apoptosis were determined using CCK-8, EdU, transwell assays and flow cytometry, respectively. Xenograft tumor experiment was performed to explore the action of matrine and AGRN on tumor growth in colorectal cancer in vivo. Immunohistochemistry (IHC) assay was applied for AGRN, ß-catenin, and c-Myc expression in the tumor tissues from mice. RESULTS: Matrine dramatically repressed cell growth and reduced the level of AGRN in colorectal cancer cells. AGRN expression was boosted colorectal cancer tissues and cells. AGRN downregulation depressed cell proliferation, migration, invasion, and enhanced cell apoptosis in colorectal cancer cells. Moreover, matrine showed the anti-tumor effects on colorectal cancer cells via regulating AGRN expression. AGRN knockdown could inactivate the Wnt/ß-catenin pathway in colorectal cancer cells. We found that AGRN downregulation exhibited the inhibition action in the progression of colorectal cancer by modulating the Wnt/ß-catenin pathway. In addition, matrine could inhibit the activation of the Wnt/ß-catenin pathway through regulating AGRN in colorectal cancer cells. Furthermore, xenograft tumor experiment revealed that matrine treatment or AGRN knockdown repressed the development of colorectal cancer via the Wnt/ß-catenin pathway in vivo. CONCLUSION: Matrine retarded colorectal cancer development by modulating AGRN to inactivate the Wnt/ß-catenin pathway.

Nuclear receptor subfamily 3 group c member 2 (NR3C2) is downregulated due to hypermethylation and plays a tumor-suppressive role in colon cancer.

Nuclear receptor subfamily 3 group c member 2 (NR3C2) has been reported to function as a tumor suppressor in several tumors. However, the clinical significance and potential action mechanisms of NR3C2 in colon cancer (COAD) remain unclear. NR3C2 expression and its correlation with clinicopathological features in COAD were analyzed based on the Cancer Genome Atlas (TCGA) and Gene Expression Omnibus (GEO) databases. Receiver operating characteristic (ROC) curves and Human Protein Atlas (HPA) database were used to evaluate the diagnostic and prognostic values of NR3C2 in COAD. Immune infiltration and DNA methylation analyses were performed by Gene Set Cancer Analysis (GSCA) database. NR3C2-correlated genes were identified by UALCAN database and subjected to gene ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) enrichment pathway analyses. Cell apoptosis and proliferation were evaluated using TUNEL and CCK-8 assays, respectively. NR3C2 was downregulated in COAD based on TCGA and GEO databases, which may be due to promoter hypermethylation. NR3C2 expression was correlated with prognosis and immune infiltration of COAD. High NR3C2 expression displayed good diagnostic value in COAD. KEGG pathway analysis presented that NR3C2-correlated genes were mainly clustered in choline metabolism in cancer and apoptosis. In vitro experiments confirmed that NR3C2 overexpression induced apoptosis and suppressed proliferation in COAD cells. In conclusion, our study revealed the potential prognostic and diagnostic values of NR3C2 and provided insights into understanding the tumor-suppressive role of NR3C2 in COAD progression.

Tumor suppressor LHX6 upregulation contributes to the inhibitory effect of miR-346 knockdown on colorectal cancer cell growth.

Colorectal cancer (CRC) is one of the prevalent types of human malignancies and ranks as the second leading cause of cancer-associated death worldwide. Dysregulated miRNAs have been promulgated as oncogenes or tumor-suppressive genes participating in the initiation and progression of CRC. A recent study reported that miR-346 was highly expressed in CRC patients. However, the biological role and underlying mechanism of miR-346 in CRC remain elusive. qRT-PCR and western blot assays were employed to detect miR-346 and LIM homeobox domain 6 (LHX6) expression in CRC cells. Cell proliferation was evaluated by CCK-8 and BrdU assays. Apoptosis was evaluated by TUNEL assay. The interaction between miR-346 and LHX6 was assessed by luciferase reporter assay. Results showed that miR-346 expression was increased and LHX6 expression was reduced in CRC cells. miR-346 knockdown and LHX6 overexpression inhibited proliferation and promoted apoptosis of CRC cells. Additionally, we found that miR-346 negatively regulated LHX6 expression in CRC cells by directly targeting LHX6. LHX6 knockdown partially attenuated anti-miR-346-induced proliferation reduction and apoptosis promotion in CRC cells. Furthermore, miR-346 knockdown inhibited the protein kinase B (Akt)/mechanistic target of rapamycin (mTOR) pathway in CRC cells by targeting LHX6. The present study indicated that miR-346 knockdown repressed cell growth in CRC cells by upregulating LHX6, and this was associated with inactivation of the Akt/mTOR pathway.

Vision Transformer-based recognition of diabetic retinopathy grade.

BACKGROUND: In the domain of natural language processing, Transformers are recognized as state-of-the-art models, which opposing to typical convolutional neural networks (CNNs) do not rely on convolution layers. Instead, Transformers employ multi-head attention mechanisms as the main building block to capture long-range contextual relations between image pixels. Recently, CNNs dominated the deep learning solutions for diabetic retinopathy grade recognition. However, spurred by the advantages of Transformers, we propose a Transformer-based method that is appropriate for recognizing the grade of diabetic retinopathy. PURPOSE: The purposes of this work are to demonstrate that (i) the pure attention mechanism is suitable for diabetic retinopathy grade recognition and (ii) Transformers can replace traditional CNNs for diabetic retinopathy grade recognition. METHODS: This paper proposes a Vision Transformer-based method to recognize the grade of diabetic retinopathy. Fundus images are subdivided into non-overlapping patches, which are then converted into sequences by flattening, and undergo a linear and positional embedding process to preserve positional information. Then, the generated sequence is input into several multi-head attention layers to generate the final representation. The first token sequence is input to a softmax classification layer to produce the recognition output in the classification stage. RESULTS: The dataset for training and testing employs fundus images of different resolutions, subdivided into patches. We challenge our method against current CNNs and extreme learning machines and achieve an appealing performance. Specifically, the suggested deep learning architecture attains an accuracy of 91.4%, specificity = 0.977 (95% confidence interval (CI) (0.951-1)), precision = 0.928 (95% CI (0.852-1)), sensitivity = 0.926 (95% CI (0.863-0.989)), quadratic weighted kappa score = 0.935, and area under curve (AUC) = 0.986. CONCLUSION: Our comparative experiments against current methods conclude that our model is competitive and highlight that an attention mechanism based on a Vision Transformer model is promising for the diabetic retinopathy grade recognition task.

Set-valued data collection with local differential privacy based on category hierarchy.

Set-valued data is extremely important and widely used in sensor technology and application. Recently, privacy protection for set-valued data under differential privacy (DP) has become a research hotspot. However, the DP model assumes that the data center is trustworthy, consequently, increasingly attention has been paid to the application of the local differential privacy model (LDP) for set-valued data. Constrained by the local differential privacy model, most methods randomly respond to the subset of set-valued data, and the data collector conducts statistics on the received data. There are two main problems with this kind of method: one is that the utility function used in the random response loses too much information; the other is that the privacy protection of the set-valued data category is usually ignored. To solve these problems, this paper proposes a set-valued data collection method (SetLDP) based on the category hierarchy under the local differential privacy model. The core idea is to first make a random response to the existence of the category, continue to disturb the item count if the category exists, and finally randomly respond to a candidate itemset based on the new utility function. Theory analysis and experimental results show that the SetLDP can not only preserve more information, but also protect the category private information in set-valued data.