Providing insights into health data science education through artificial intelligence.

BACKGROUND: Health Data Science (HDS) is a novel interdisciplinary field that integrates biological, clinical, and computational sciences with the aim of analysing clinical and biological data through the utilisation of computational methods. Training healthcare specialists who are knowledgeable in both health and data sciences is highly required, important, and challenging. Therefore, it is essential to analyse students' learning experiences through artificial intelligence techniques in order to provide both teachers and learners with insights about effective learning strategies and to improve existing HDS course designs. METHODS: We applied artificial intelligence methods to uncover learning tactics and strategies employed by students in an HDS massive open online course with over 3,000 students enrolled. We also used statistical tests to explore students' engagement with different resources (such as reading materials and lecture videos) and their level of engagement with various HDS topics. RESULTS: We found that students in HDS employed four learning tactics, such as actively connecting new information to their prior knowledge, taking assessments and practising programming to evaluate their understanding, collaborating with their classmates, and repeating information to memorise. Based on the employed tactics, we also found three types of learning strategies, including low engagement (Surface learners), moderate engagement (Strategic learners), and high engagement (Deep learners), which are in line with well-known educational theories. The results indicate that successful students allocate more time to practical topics, such as projects and discussions, make connections among concepts, and employ peer learning. CONCLUSIONS: We applied artificial intelligence techniques to provide new insights into HDS education. Based on the findings, we provide pedagogical suggestions not only for course designers but also for teachers and learners that have the potential to improve the learning experience of HDS students.

DisCoVering potential candidates of RNAi-based therapy for COVID-19 using computational methods.

The ongoing pandemic of a novel coronavirus (SARS-CoV-2) leads to international concern; thus, emergency interventions need to be taken. Due to the time-consuming experimental methods for proposing useful treatments, computational approaches facilitate investigating thousands of alternatives simultaneously and narrow down the cases for experimental validation. Herein, we conducted four independent analyses for RNA interference (RNAi)-based therapy with computational and bioinformatic methods. The aim is to target the evolutionarily conserved regions in the SARS-CoV-2 genome in order to down-regulate or silence its RNA. miRNAs are denoted to play an important role in the resistance of some species to viral infections. A comprehensive analysis of the miRNAs available in the body of humans, as well as the miRNAs in bats and many other species, were done to find efficient candidates with low side effects in the human body. Moreover, the evolutionarily conserved regions in the SARS-CoV-2 genome were considered for designing novel significant siRNA that are target-specific. A small set of miRNAs and five siRNAs were suggested as the possible efficient candidates with a high affinity to the SARS-CoV-2 genome and low side effects. The suggested candidates are promising therapeutics for the experimental evaluations and may speed up the procedure of treatment design. Materials and implementations are available at: https://github.com/nrohani/SARS-CoV-2.

Classifying Breast Cancer Molecular Subtypes by Using Deep Clustering Approach.

Cancer is a complex disease with a high rate of mortality. The characteristics of tumor masses are very heterogeneous; thus, the appropriate classification of tumors is a critical point in the effective treatment. A high level of heterogeneity has also been observed in breast cancer. Therefore, detecting the molecular subtypes of this disease is an essential issue for medicine that could be facilitated using bioinformatics. This study aims to discover the molecular subtypes of breast cancer using somatic mutation profiles of tumors. Nonetheless, the somatic mutation profiles are very sparse. Therefore, a network propagation method is used in the gene interaction network to make the mutation profiles dense. Afterward, the deep embedded clustering (DEC) method is used to classify the breast tumors into four subtypes. In the next step, gene signature of each subtype is obtained using Fisher's exact test. Besides the enrichment of gene signatures in numerous biological databases, clinical and molecular analyses verify that the proposed method using mutation profiles can efficiently detect the molecular subtypes of breast cancer. Finally, a supervised classifier is trained based on the discovered subtypes to predict the molecular subtype of a new patient. The code and material of the method are available at: https://github.com/nrohani/MolecularSubtypes.

Drug-Drug Interaction Predicting by Neural Network Using Integrated Similarity.

Drug-Drug Interaction (DDI) prediction is one of the most critical issues in drug development and health. Proposing appropriate computational methods for predicting unknown DDI with high precision is challenging. We proposed "NDD: Neural network-based method for drug-drug interaction prediction" for predicting unknown DDIs using various information about drugs. Multiple drug similarities based on drug substructure, target, side effect, off-label side effect, pathway, transporter, and indication data are calculated. At first, NDD uses a heuristic similarity selection process and then integrates the selected similarities with a nonlinear similarity fusion method to achieve high-level features. Afterward, it uses a neural network for interaction prediction. The similarity selection and similarity integration parts of NDD have been proposed in previous studies of other problems. Our novelty is to combine these parts with new neural network architecture and apply these approaches in the context of DDI prediction. We compared NDD with six machine learning classifiers and six state-of-the-art graph-based methods on three benchmark datasets. NDD achieved superior performance in cross-validation with AUPR ranging from 0.830 to 0.947, AUC from 0.954 to 0.994 and F-measure from 0.772 to 0.902. Moreover, cumulative evidence in case studies on numerous drug pairs, further confirm the ability of NDD to predict unknown DDIs. The evaluations corroborate that NDD is an efficient method for predicting unknown DDIs. The data and implementation of NDD are available at https://github.com/nrohani/NDD .