In this research, we present a deep learning model with k-mer as the only feature for classifying lncRNAs and mRNAs from RNA sequences alone. We compared the model with PLEK, CPC, PlncRNA-HDeep, and CICN using human datasets. The findings demonstrated that our model has an accuracy of up to 98.2%, indicating that it is superior to all of the aforementioned techniques.
In addition, this study proposes five other Machine Learning algorithms, namely Naive Bayes (NB), AdaBoost, K-Nearest Neighbor (KNN), support vector machine and Random Forest(RF) Classifiers, to predict lncRNAs from RNA-seq data. The accuracy of these models is compared to that of our deep learning model, and the best model for classification is then recommended.
Funding: This research was funded by Ministry of Environment, Water and Agriculture of Saudi Arabia