Computational Biology and Chemistry最新文献

In the realm of plant biology, understanding the intricate regulatory mechanisms governing stress responses stands as a pivotal pursuit. Circular RNAs (circRNAs), emerging as critical players in gene regulation, have garnered attention in recent days for their potential roles in abiotic stress adaptation. A comprehensive grasp of circRNAs' functions in stress response offers avenues for breeders to manipulating plants to develop abiotic stress resistant crop cultivars to thrive in challenging climates. This study pioneers a machine learning-based model for predicting abiotic stress-responsive circRNAs. The K-tuple nucleotide composition (KNC) and Pseudo KNC (PKNC) features were utilized to numerically represent circRNAs. Three different feature selection strategies were employed to select relevant and non-redundant features. Eight shallow and four deep learning algorithms were evaluated to build the final predictive model. Following five-fold cross-validation process, XGBoost learning algorithm demonstrated superior performance with LightGBM-chosen 260 KNC features (Accuracy: 74.55 %, auROC: 81.23 %, auPRC: 76.52 %) and 160 PKNC features (Accuracy: 74.32 %, auROC: 81.04 %, auPRC: 76.43 %), over other combinations of learning algorithms and feature selection techniques. Further, the robustness of the developed models were evaluated using an independent test dataset, where the overall accuracy, auROC and auPRC were found to be 73.13 %, 72.34 % and 72.68 % for KNC feature set and 73.52 %, 79.53 % and 73.09 % for PKNC feature set, respectively. This computational approach was also integrated into an online prediction tool, AScirRNA (https://iasri-sg.icar.gov.in/ascirna/) for easy prediction by the users. Both the proposed model and the developed tool are poised to augment ongoing efforts in identifying stress-responsive circRNAs in plants.

Numerous studies have shown that microRNAs (miRNAs) play a key role in human diseases as critical biomarkers. Its abnormal expression is often accompanied by the emergence of specific diseases. Therefore, studying the relationship between miRNAs and diseases can deepen the insights of their pathogenesis, grasp the process of disease onset and development, and promote drug research of specific diseases. However, many undiscovered relationships between miRNAs and diseases remain, significantly limiting research on miRNA-disease correlations. To explore more potential correlations, we propose a dual-channel convolutional model based on stacked deep learning collaborative gradient decomposition for predicting miRNA-disease associations (DCSGMDA). Firstly, we constructed similarity networks for miRNAs and diseases, as well as an association relationship network. Secondly, potential features were fully mined using stacked deep learning and gradient decomposition networks, along with dual-channel convolutional neural networks. Finally, correlations were scored by a multilayer perceptron. We performed 5-fold and 10-fold cross-validation experiments on DCSGMDA using two datasets based on the Human MicroRNA Disease Database (HMDD). Additionally, parametric, ablation, and comparative experiments, along with case studies, were conducted. The experimental results demonstrate that DCSGMDA performs well in predicting miRNA-disease associations.

Multiple types of omics data contain a wealth of biomedical information which reflect different aspects of clinical samples. Multi-omics integrated analysis is more likely to lead to more accurate clinical decisions. Existing cancer diagnostic methods based on multi-omics data integration mainly focus on the classification accuracy of the model, while neglecting the interpretability of the internal mechanism and the reliability of the results, which are crucial in specific domains such as precision medicine and the life sciences. To overcome this limitation, we propose a trustworthy multi-omics dynamic learning framework (TMODINET) for cancer diagnostic. The framework employs multi-omics adaptive dynamic learning to process each sample to provide patient-centered personality diagnosis by using self-attentional learning of features and modalities. To characterize the correlation between samples well, we introduce a graph dynamic learning method which can adaptively adjust the graph structure according to the specific classification results for specific graph convolutional networks (GCN) learning. Moreover, we utilize an uncertainty mechanism by employing Dirichlet distribution and Dempster–Shafer theory to obtain uncertainty and integrate multi-omics data at the decision level, ensuring trustworthy for cancer diagnosis. Extensive experiments on four real-world multimodal medical datasets are conducted. Compared to state-of-the-art methods, the superior performance and trustworthiness of our proposed algorithm are clearly validated. Our model has great potential for clinical diagnosis.

Objective:

The prediction of sepsis, especially early diagnosis, has received a significant attention in biomedical research. In order to improve current medical scoring system and overcome the limitations of class imbalance and sample size of local EHR (electronic health records), we propose a novel knowledge-transfer-based approach, which combines a medical scoring system and an ordinal logistic regression model.

Materials and Methods:

Medical scoring systems (i.e. NEWS, SIRS and QSOFA) are generally robust and useful for sepsis diagnosis. With local EHR, machine-learning-based methods have been widely used for building prediction models/methods, but they are often impacted by class imbalance and sample size. Knowledge distillation and knowledge transfer have recently been proposed as a combination approach for improving the prediction performance and model generalization. In this study, we developed a novel knowledge-transfer-based method for combining a medical scoring system (after a proposed score transformation) and an ordinal logistic regression model. We mathematically confirmed that it was equivalent to a specific form of the weighted regression. Furthermore, we theoretically explored its effectiveness in the scenario of class imbalance.

Results:

For the local dataset and the MIMIC-IV dataset, the VUS (the volume under the multi-dimensional ROC surface, a generalization measure of AUC-ROC for ordinal categories) of the knowledge-transfer-based model (ORNEWS) based on the NEWS scoring system were 0.384 and 0.339, respectively, while the VUS of the traditional ordinal regression model (OR) were 0.352 and 0.322, respectively. Consistent analysis results were also observed for the knowledge-transfer-based models based on the SIRS/QSOFA scoring systems in the ordinal scenarios. Additionally, the predicted probabilities and the binary classification ROC curves of the knowledge-transfer-based models indicated that this approach enhanced the predicted probabilities for the minority classes while reducing the predicted probabilities for the majority classes, which improved AUCs/VUSs on imbalanced data.

Discussion:

Knowledge transfer, which combines a medical scoring system and a machine-learning-based model, improves the prediction performance for early diagnosis of sepsis, especially in the scenarios of class imbalance and limited sample size.

According to the considered role of lipophilicity-hydrophobicity on organic Schiff base hydrazones, different substituents of phenyl, ethyl, and methyl groups were inserted in the synthetic strategy of diisatin dihydrazones (L1–4). The biochemical enhancement was evaluated depending on their inhibitive potential of the growth power of three human tumor cells, fungi, and bacteria. The biochemical assays assigned the effected role of different substituents of phenyl, ethyl, and methyl groups on the effectiveness of their diisatin dihydrazone reagents. The interacting modes with calf thymus DNA (i.e. Ct-DNA) were studied via viscometric and spectrophotometric titration.

The organo-reagent L1 with the oxalic derivative assigned a performed inhibitive action for the examined microbes and the human tumor cell lines growing up over the terephthalic (L4) > malonic (L2) > succinic (L3) ones. From K_b = binding constant, and $∆G_b^{\ne }$ = Gibb’s free energy values, the binding of interaction within Ct-DNA was evaluated for all compounds (L1–4), in which L1, L3, and L4 assigned the highest reactivity referring to the covalent/non-covalent modes of interaction, as given for (L1–4), 14.32, 13.28, 10.87, and 12.41 × 10⁷ mol⁻¹ dm³, and −45^.17, −43.24, −43.75, and −44.05 kJ mol⁻¹, respectively. DFT and docking studies were achieved to support the current work.

Ferroptosis is a newly identified form of non-apoptotic programmed cell death resulting from iron-dependent lipid peroxidation. It is controlled by integrated oxidation and antioxidant systems. Ferroptosis exerts a crucial effect on the carcinogenesis of several cancers, including pulmonary cancer. Herein, a ferroptosis-associated gene signature for lung cancer prognosis and diagnosis was identified using integrative bioinformatics analyses. From the FerrDB database, 256 ferroptotic regulators and markers were identified. Of these, 25 exhibited differential expression between lung cancer and non-cancerous samples, as evidenced by the GSE19804 and GSE7670 datasets from the GEO database. Utilizing LASSO Cox regression analysis on TCGA-LUAD data, a potent 3-gene risk signature comprising CAV1, RRM2, and EGFR was established. This signature adeptly differentiates various survival outcomes in lung cancer patients, including overall survival and disease-specific intervals. Based on the 3-gene risk signature, lung cancer patients were categorized into high-risk and low-risk groups. Comparative analysis revealed 69 differentially expressed genes between these groups, with UBE2Z significantly associated with overall survival in TCGA-LUAD. UBE2Z was found to be upregulated in LUAD tissues and cells compared to normal controls. Functionally, the knockdown of UBE2Z curtailed aggressive behaviors in LUAD cells, including viability, migration, and invasion. Moreover, this knockdown led to a decrease in the mesenchymal marker vimentin while elevating the epithelial marker E-cadherin within LUAD cell lines. In conclusion, the ferroptosis-associated 3-gene risk signature effectively differentiates prognosis and clinical features in patients with lung cancer. UBE2Z was identified through this model, and it is upregulated in LUAD samples. Its knockdown inhibits aggressive cellular behaviors, suggesting UBE2Z's potential as a therapeutic target for lung cancer treatment.

Polycystic ovary syndrome (PCOS) is one of the most common anovulatory disorder observed in women presenting with infertility. Several high and low throughput studies on PCOS have led to accumulation of vast amount of information on PCOS. Despite the availability of several resources which index the advances in PCOS, information on its etiology still remains inadequate. Analysis of the existing information using an integrated evidence based approach may aid identification of novel potential candidate genes with a role in PCOS pathophysiology. This work focuses on integrating existing information on PCOS from literature and gene expression studies and evaluating the application of gene prioritization and network analysis to predict missing novel candidates. Further, it assesses the utility of evidence-based scoring to rank genes for their association with PCOS. The results of this study led to identification of ∼2000 plausible candidate genes associated with PCOS. Insilico validation of these identified candidates confirmed the role of 938 genes in PCOS. Further, experimental validation was carried out for four of the potential candidate genes, a high-scoring (PROS1), two mid-scoring (C1QA and KNG1), and a low-scoring gene (VTN) involved in the complement and coagulation pathway by comparing protein levels in follicular fluid in women with PCOS and healthy controls. While the expression of PROS1, C1QA, and KNG1 was found to be significantly downregulated in women with PCOS, the expression of VTN was found to be unchanged in PCOS. The findings of this study reiterate the utility of employing insilico approaches to identify and prioritize the most promising candidate genes in diseases with a complex pathophysiology like PCOS. Further, the study also helps in gaining clearer insights into the molecular mechanisms associated with the manifestation of the PCOS phenotype by contributing to the existing repertoire of genes associated with PCOS.

Autism Spectrum Disorder (ASD) is a neurological illness that degrades communication and interaction among others. Autism can be detected at any stage. Early detection of ASD is important in preventing the communication, interaction and behavioral outcomes of individuals. Hence, this research introduced the Fractional Whale-driving Driving Training-based Based Optimization with Convolutional Neural Network-based Transfer learning (FWDTBO-CNN_TL) for identifying ASD. Here, the FWDTBO is modelled by the incorporation of Fractional calculus (FC), Whale optimization algorithm (WOA) and Driving Training-based Optimization (DTBO) that trains the hyperparameters of CNN-TL. Moreover, the Convolutional Neural Networks (CNN) utilize the hyperparameters from trained models, like Alex Net and Shuffle Net in such a way that the CNN-TL is designed. To improve the detection efficiency, the nub region was extracted and carried out with the functional connectivity-based Whale Driving Training Optimization (WDTBO) algorithm. Moreover, the TL is tuned by the FWDTBO algorithm. The result reveals that the ASD detection technique, FWDTBO-CNN-TL acquired 90.7 % accuracy, 95.4 % sensitivity, 93.7 % specificity and 93 % f-measure with the ABIDE-II dataset.

Antimetabolites are the most effective chemotherapeutics for treating cancer. They have exerted their anticancer effects by interfering with DNA synthesis. Recently, interest in modified nucleoside analogues has grown due to their superior efficiency. Nucleoside analogue derivatives have emerged as crucial candidates for cancer treatment due to their ability to target the cells responsible for cancer within the body specifically. The ability of nucleoside analogues derivatives to target specific molecular pathways has reduced toxicity and increased efficiency compared to traditional chemotherapy drugs. Nucleoside analogues have interfered with physiological nucleosides and induced cytotoxicity in cancerous cells. In this investigation, derivatives of ribofuranose nucleoside analogues have been designed. Density functional theory (DFT) calculations have been performed at the B3LYP/6–311 G(d,p) level. The designed molecules have been characterized by UV/Vis spectroscopy using the CPCM model in DMSO solvent, and molecular structural parameters, such as HOMO/LUMO and MEPS, have been determined. Derivative d1m has exhibited a high energy gap and low absorption energy compared to the other derivatives. Molecular docking of the designed molecules (d1o-d2m) has been performed with the EGFR and VEGFR2 proteins. d2o has shown good binding energy with the EGFR protein, while d1o has shown good results with VEGFR2. Global chemical parameters and NBO analysis have been conducted to investigate the derivatives charge transfer properties and chemical reactivity. NBO analysis has provided information about the donor and acceptor parts within a molecule, while global chemical parameters have given insights into the reactivity, stability, and solubility of the molecules. It has been found that the derivatives are more chemically reactive, thermodynamically stable, and have better binding affinity than the parent molecule. Based on the analysis, the drug interaction with the cancer-causing protein makes it more effective for cancer treatment.

Ficus viren has been traditionally used to treat diabetes, and its extract inhibits carbohydrate/lipid metabolism and possesses anti-hyperglycemic potential. However, there is conflicting investigation related to F. viren extract effect on carbohydrate metabolism. Thus, bioactive and mechanism behind its antidiabetic potential is still scanty. This study explored F. viren’s anti-diabetic property by identifying potential phytoconstituents and mechanism. A sequential in-silico approach was used i.e., druglikeness, molecular docking, post-docking MM-GBSA, ADMET studies, molecular dynamic simulation (MDS), and post-MDS MM-GBSA. We screened ∼32 phytoconstituents and twelve potential organ-specific diabetic targets (O.S.D.Ts i.e., IR, DPP-4, ppar-γ, ppar-α, ppar-δ, GLP-1R, SIRT-1, AMPK, GSK-3β, RAGE, and AR). Drug likeness study identified 18 druggable candidates among 32 phytoconstituents. K3A, quercetin, scutellarein, sorbifolin, and vogeline J identified as potential ligands from druggable ligands, using IR as the standard target. Subsequently, potential ligands docked with remaining O.S.D.Ts. and data showed that K3A binds strongly with AMPK, ppar-δ, DPP-4, and GSK-3β, while scutellarein binds with AR and ppar-α. Sorbifolin, quercetin, and vogeline J binds with ppar-α, ppar-γ, and RAGE, respectively. Post-docking MM-GBSA data (∆G_Bind) also depicted potential ligand’s strong binding affinities with their corresponding targets. Thereafter, simulation data revealed that only scutellarein and sorbifolin showed dynamic stability with their respective targets, i.e., AR/ppar-α and ppar-α, respectively. Interestingly, post-MDS MM-GBSA revealed that only scutellarein exhibited strong ∆G_Bind of −55.08 kcal/mol and −75.48 kcal/mol with AR and ppar-α, respectively. Though, collective computational analysis supports antidiabetic potential of F. viren through AR and ppar-α modulation by scutellarein.