NN-RNALoc: Neural network-based model for prediction of mRNA sub-cellular localization using distance-based sub-sequence profiles.

The localization of messenger RNAs (mRNAs) is a frequently observed phenomenon and a crucial aspect of gene expression regulation. It is also a mechanism for targeting proteins to a specific cellular region. Moreover, prior research and studies have shown the significance of intracellular RNA positioning during embryonic and neural dendrite formation. Incorrect RNA localization, which can be caused by a variety of factors, such as mutations in trans-regulatory elements, has been linked to the development of certain neuromuscular diseases and cancer. In this study, we introduced NN-RNALoc, a neural network-based method for predicting the cellular location of mRNA using novel features extracted from mRNA sequence data and protein interaction patterns. In fact, we developed a distance-based subsequence profile for RNA sequence representation that is more memory and time-efficient than well-known k-mer sequence representation. Combining protein-protein interaction data, which is essential for numerous biological processes, with our novel distance-based subsequence profiles of mRNA sequences produces more accurate features. On two benchmark datasets, CeFra-Seq and RNALocate, the performance of NN-RNALoc is compared to powerful predictive models proposed in previous works (mRNALoc, RNATracker, mLoc-mRNA, DM3Loc, iLoc-mRNA, and EL-RMLocNet), and a ground neural (DNN5-mer) network. Compared to the previous methods, NN-RNALoc significantly reduces computation time and also outperforms them in terms of accuracy. This study's source code and datasets are freely accessible at https://github.com/NeginBabaiha/NN-RNALoc.

Identification of repurposed drugs targeting significant long non-coding RNAs in the cross-talk between diabetes mellitus and Alzheimer's disease.

The relationship between diabetes mellitus (DM) and Alzheimer's disease (AD) is so strong that scientists called it "brain diabetes". According to several studies, the critical factor in this relationship is brain insulin resistance. Due to the rapid global spread of both diseases, overcoming this cross-talk has a significant impact on societies. Long non-coding RNAs (lncRNAs), on the other hand, have a substantial impact on complex diseases due to their ability to influence gene expression via a variety of mechanisms. Consequently, the regulation of lncRNA expression in chronic diseases permits the development of innovative therapeutic techniques. However, developing a new drug requires considerable time and money. Recently repurposing existing drugs has gained popularity due to the use of low-risk compounds, which may result in cost and time savings. in this study, we identified drug repurposing candidates capable of controlling the expression of common lncRNAs in the cross-talk between DM and AD. We also utilized drugs that interfered with this cross-talk. To do this, high degree common lncRNAs were extracted from microRNA-lncRNA bipartite network. The drugs that interact with the specified lncRNAs were then collected from multiple data sources. These drugs, referred to as set D, were classified in to positive (D+) and negative (D-) groups based on their effects on the expression of the interacting lncRNAs. A feature selection algorithm was used to select six important features for D. Using a random forest classifier, these features were capable of classifying D+ and D- with an accuracy of 82.5%. Finally, the same six features were extracted for the most recently Food and Drug Administration (FDA) approved drugs in order to identify those with the highest likelihood of belonging to D+ or D-. The most significant FDA-approved positive drugs, chromium nicotinate and tapentadol, were presented as repurposing candidates, while cefepime and dihydro-alpha-ergocryptine were recommended as significant adverse drugs. Moreover, two natural compounds, curcumin and quercetin, were recommended to prevent this cross-talk. According to the previous studies, less attention has been paid to the role of lncRNAs in this cross-talk. Our research not only did identify important lncRNAs, but it also suggested potential repurposed drugs to control them.

Exploring the role of non-coding RNAs as potential candidate biomarkers in the cross-talk between diabetes mellitus and Alzheimer's disease.

Background: Recent research has investigated the connection between Diabetes Mellitus (DM) and Alzheimer's Disease (AD). Insulin resistance plays a crucial role in this interaction. Studies have focused on dysregulated proteins to disrupt this connection. Non-coding RNAs (ncRNAs), on the other hand, play an important role in the development of many diseases. They encode the majority of the human genome and regulate gene expression through a variety of mechanisms. Consequently, identifying significant ncRNAs and utilizing them as biomarkers could facilitate the early detection of this cross-talk. On the other hand, computational-based methods may help to understand the possible relationships between different molecules and conduct future wet laboratory experiments. Materials and methods: In this study, we retrieved Genome-Wide Association Study (GWAS, 2008) results from the United Kingdom Biobank database using the keywords "Alzheimer's" and "Diabetes Mellitus." After excluding low confidence variants, statistical analysis was performed, and adjusted p-values were determined. Using the Linkage Disequilibrium method, 127 significant shared Single Nucleotide Polymorphism (SNP) were chosen and the SNP-SNP interaction network was built. From this network, dense subgraphs were extracted as signatures. By mapping each signature to the reference genome, genes associated with the selected SNPs were retrieved. Then, protein-microRNA (miRNA) and miRNA-long non-coding RNA (lncRNA) bipartite networks were built and significant ncRNAs were extracted. After the validation process, by applying the scoring function, the final protein-miRNA-lncRNA tripartite network was constructed, and significant miRNAs and lncRNAs were identified. Results: Hsa-miR-199a-5p, hsa-miR-199b-5p, hsa-miR-423-5p, and hsa-miR-3184-5p, the four most significant miRNAs, as well as NEAT1, XIST, and KCNQ1OT1, the three most important lncRNAs, and their interacting proteins in the final tripartite network, have been proposed as new candidate biomarkers in the cross-talk between DM and AD. The literature review also validates the obtained ncRNAs. In addition, miRNA/lncRNA pairs; hsa-miR-124-3p/KCNQ1OT1, hsa-miR-124-3p/NEAT1, and hsa-miR-124-3p/XIST, all expressed in the brain, and their interacting proteins in our final network are suggested for future research investigation. Conclusion: This study identified 127 shared SNPs, 7 proteins, 15 miRNAs, and 11 lncRNAs involved in the cross-talk between DM and AD. Different network analysis and scoring function suggested the most significant miRNAs and lncRNAs as potential candidate biomarkers for wet laboratory experiments. Considering these candidate biomarkers may help in the early detection of DM and AD co-occurrence.