i4mC-Deep: An Intelligent Predictor of N4-Methylcytosine Sites Using a Deep Learning Approach with Chemical Properties

Abstract

DNA is subject to epigenetic modification by the molecule N4-methylcytosine (4mC). N4-methylcytosine plays a crucial role in DNA repair and replication, protects host DNA from degradation, and regulates DNA expression. However, though current experimental techniques can identify 4mC sites, such techniques are expensive and laborious. Therefore, computational tools that can predict 4mC sites would be very useful for understanding the biological mechanism of this vital type of DNA modification. Conventional machine-learning-based methods rely on hand-crafted features, but the new method saves time and computational cost by making use of learned features instead. In this study, we propose i4mC-Deep, an intelligent predictor based on a convolutional neural network (CNN) that predicts 4mC modification sites in DNA samples. The CNN is capable of automatically extracting important features from input samples during training. Nucleotide chemical properties and nucleotide density, which together represent a DNA sequence, act as CNN input data. The outcome of the proposed method outperforms several state-of-the-art predictors. When i4mC-Deep was used to analyze G. subterruneus DNA, the accuracy of the results was improved by 3.9% and MCC increased by 10.5% compared to a conventional predictor.

iProm-Sigma54: A CNN Base Prediction Tool for σ54 Promoters

“…Thus, this study utilized a one-hot feature encoding scheme. Several recent cutting edge bioinformatics techniques have used this technique e [ 40 , 41 , 42 , 43 ]. Representation of each nucleotide for A , C , G , and T , characterized as follows: …”

Predicting the bandgap and efficiency of perovskite solar cells using machine learning methods

“…Based on the characteristics of the materials used and the design of the devices, machine learning (ML) has emerged as an effective method for predicting the performance of PSCs. Large databases of experimental and theoretical data can be analyzed using ML algorithms to identify patterns and correlations that can be used to create prediction models [14][15][16][17][18][19][20][21][22][23]. By training ML models on data from high-throughput experimental and theoretical studies, it is possible to accelerate the development of new PSC materials and optimize device performance.…”

Attention-Based Graph Neural Network for Molecular Solubility Prediction

“…Traditional analytical techniques fail to process large amounts of data; therefore, it is necessary to process these data and transform them into knowledge . Owing to the large amount of data generated during DD, it is possible to understand the properties and actions that are beneficial for drug design. − Hence, presently, designing machine learning techniques that can train on molecular big data and predict molecular properties is of great interest. Early computational solubility prediction efforts, based on molecular structure, focused primarily on developing regression models to predict solubility, using the structural and electronic properties of the molecules as inputs. − Many machine learning techniques have been reported in the literature for predicting molecular properties; − the most common is molecular graph representation.…”