ASRmiRNA: Abiotic Stress-Responsive miRNA Prediction in Plants by Using Machine Learning Algorithms with Pseudo K-Tuple Nucleotide Compositional Features

MicroRNAs (miRNAs) play a significant role in plant response to different abiotic stresses. Thus, identification of abiotic stress-responsive miRNAs holds immense importance in crop breeding programmes to develop cultivars resistant to abiotic stresses. In this study, we developed a machine learning-based computational method for prediction of miRNAs associated with abiotic stresses. Three types of datasets were used for prediction, i.e., miRNA, Pre-miRNA, and Pre-miRNA + miRNA. The pseudo K-tuple nucleotide compositional features were generated for each sequence to transform the sequence data into numeric feature vectors. Support vector machine (SVM) was employed for prediction. The area under receiver operating characteristics curve (auROC) of 70.21, 69.71, 77.94 and area under precision-recall curve (auPRC) of 69.96, 65.64, 77.32 percentages were obtained for miRNA, Pre-miRNA, and Pre-miRNA + miRNA datasets, respectively. Overall prediction accuracies for the independent test set were 62.33, 64.85, 69.21 percentages, respectively, for the three datasets. The SVM also achieved higher accuracy than other learning methods such as random forest, extreme gradient boosting, and adaptive boosting. To implement our method with ease, an online prediction server “ASRmiRNA” has been developed. The proposed approach is believed to supplement the existing effort for identification of abiotic stress-responsive miRNAs and Pre-miRNAs.     

Circulating miRNA Expression Profiles and Machine Learning Models in Association with Response to Irinotecan-Based Treatment in Metastatic Colorectal Cancer

Colorectal cancer represents a leading cause of cancer-related morbidity and mortality. Despite improvements, chemotherapy remains the backbone of colorectal cancer treatment. The aim of this study is to investigate the variation of circulating microRNA expression profiles and the response to irinotecan-based treatment in metastatic colorectal cancer and to identify relevant target genes and molecular functions. Serum samples from 95 metastatic colorectal cancer patients were analyzed. The microRNA expression was tested with a NucleoSpin miRNA kit (Machnery-Nagel, Germany), and a machine learning approach was subsequently applied for microRNA profiling. The top 10 upregulated microRNAs in the non-responders group were hsa-miR-181b-5p, hsa-miR-10b-5p, hsa-let-7f-5p, hsa-miR-181a-5p, hsa-miR-181d-5p, hsa-miR-301a-3p, hsa-miR-92a-3p, hsa-miR-155-5p, hsa-miR-30c-5p, and hsa-let-7i-5p. Similarly, the top 10 downregulated microRNAs were hsa-let-7d-5p, hsa-let-7c-5p, hsa-miR-215-5p, hsa-miR-143-3p, hsa-let-7a-5p, hsa-miR-10a-5p, hsa-miR-142-5p, hsa-miR-148a-3p, hsa-miR-122-5p, and hsa-miR-17-5p. The upregulation of microRNAs in the miR-181 family and the downregulation of those in the let-7 family appear to be mostly involved with non-responsiveness to irinotecan-based treatment.     

Utility of a Molecular Signature for Predicting Recurrence and Progression in Non-Muscle-Invasive Bladder Cancer Patients: Comparison with the EORTC,CUETO and 2021 EAU Risk Groups

To evaluate the utility of different risk assessments in non-muscle-invasive bladder cancer (NMIBC) patients, a total of 178 NMIBC patients from Chungbuk National University Hospital (CBNUH) were enrolled, and the predictive value of the molecular signature-based subtype predictor (MSP888) and risk calculators based on clinicopathological factors (EORTC, CUETO and 2021 EAU risk scores) was compared. Of the 178 patients, 49 were newly analyzed by the RNA-sequencing, and their MSP888 subtype was evaluated. The ability of the EORTC, MSP888 and two molecular subtyping systems of bladder cancer (Lund and UROMOL subtypes) to predict progression of 460 NMIBC patients from the UROMOL project was assessed. Cox regression analyses showed that the MSP888 was an independent predictor of NMIBC progression in the CBNUH cohort (p = 0.043). Particularly in patients without an intravesical BCG immunotherapy, MSP888 significantly linked with risk of disease recurrence and progression (both p < 0.05). However, the EORTC, CUETO and 2021 EAU risk scores showed disappointing results with respect to estimating the NMIBC prognosis. In the UROMOL cohort, the MSP888, Lund and UROMOL subtypes demonstrated a similar capacity to predict NMIBC progression (all p < 0.05). Conclusively, the MSP888 is favorable for stratifying patients to facilitate optimal treatment.