The importance of gene expression data processing in solving the classification task is determined by its ability to discern intricate patterns and relationships within genetic information, enabling the precise categorization and understanding of various gene expression profiles and their consequential impacts on biological processes and traits. In this study, we investigated various architectures and types of recurrent neural networks focusing on gene expression data. The effectiveness of the appropriate model was evaluated using various classification quality criteria based on type 1 and type 2 errors. Moreover, we calculated the integrated F1-score index using the Harrington desirability method, the value of which allowed us to improve the objectivity of the decision making when model effectiveness was evaluated. The final decision regarding model effectiveness was made based on a comprehensive classification quality criterion, which was calculated as the weighted sum of classification accuracy, integrated F1-score index, and loss function values. The simulation results show higher appeal of a single-layer GRU recurrent network with 75 neurons in the recurrent layer. We also compared convolutional and recurrent neural networks on gene expression data classification. Although convolutional neural networks showcase benefits in terms of loss function value and training time, a comparative analysis revealed that in terms of classification accuracy calculated on the test data subset, the GRU neural network model is slightly better than the CNN and LSTM models. The classification accuracy when using the GRU network was 97.2%; in other cases, it was 97.1%. In the first case, 954 out of 981 objects were correctly identified. In other cases, 952 objects were correctly identified.

One of the current focuses of modern bioinformatics is the development of hybrid models to process gene expression data, in order to create diagnostic systems for various diseases. In this study, we propose a solution to this problem that combines an inductive spectral clustering algorithm, random forest classifier, convolutional neural network, and alternative voting method for making the final decision about patient condition. In the first stage, we apply the spectral clustering algorithm to gene expression profiles using inductive methods of objective clustering, with the calculation of internal, external, and balance clustering quality criteria. This results in clusters of mutually correlated and differently expressed gene expression profiles. In the second stage, we apply the random forest classifier and convolutional neural network to identify the examined objects, containing as attributes the gene expression values in the allocated clusters. The presented research solves both binary- and multi-classification tasks. The final decision about the patient’s condition is made using the alternative voting method, considering the classification results based on the gene expression data in various clusters. The simulation results showed that the proposed technique was highly effective, achieving a high accuracy in object identification when both classifiers were used. However, the convolutional neural network had a significantly higher data processing efficiency than the random forest algorithm, due to its substantially shorter processing time.

The development of hybrid models focused on gene expression data processing for the allocation of differentially expressed and mutually correlated genes is one of the current directions in modern bioinformatics. The solution to this problem can allow us to improve the effectiveness of existing systems for complex diseases diagnosis based on gene expression data analysis on the one hand and increase the efficiency of gene regulatory network reconstruction procedures by more careful selection of genes by considering the type of disease on the other hand. In this research, we propose a stepwise procedure to form the subsets of mutually correlated and differentially expressed gene expression profiles (GEP). Firstly, we allocate an informative GEP in terms of statistical and entropy criteria using the Harrington desirability function. Then, we performed cluster analysis using SOTA and spectral clustering algorithms implemented within the framework of objective clustering inductive technology. The result of this step’s implementation is a set of clusters containing co- and differentially expressed GEPs. Validation of the model was performed using a one-dimensional two-layer convolutional neural network (CNN). The analysis of the simulation results has shown the high efficiency of the proposed model. The clusters of GEPs formed based on the clustering quality criteria values allowed us to identify the investigated objects with high accuracy. Moreover, the simulation results have also shown that the hybrid inductive model based on the spectral clustering algorithm is more effective in comparison with the use of the SOTA clustering algorithm in terms of both the complexity of the formed optimal cluster structure and the classification accuracy of the objects that contain the allocated gene expression data as attributes. The proposed hybrid inductive model contributes to increasing objectivity during the formation of the subsets of differentially and co-expressed gene expression profiles for further their application in various disease diagnosis systems and for gene regulatory network reconstruction.

The manuscript details the outcomes of a comprehensive study on the application of cluster-bicluster analysis, gene ontology analysis, and convolutional neural network (CNN) for diagnosing cancer and Alzheimer’s disease using gene expression data derived from both DNA microarray experiments and mRNA sequencing. It outlines a conceptual framework and provides a block diagram of the stepwise procedure for analyzing gene expression data, aiming to enhance the accuracy and objectivity of disease diagnosis. The research methodology involves initial gene ontology analysis, followed by the application of the Self Organizing Tree Algorithm (SOTA) for clustering gene expression profiles, an ensemble algorithm for data biclustering, and CNN for sample classification. Bayesian optimization method was employed to determine the optimal hyperparameters for all models. The analysis of simulation results demonstrates the high efficacy of the proposed approach. Specifically, for Alzheimer’s data, the number of genes analyzed was reduced from 44,662 to 4,004. Subsequent cluster-bicluster analysis divided this data into two subsets containing 1,158 and 2,846 genes, respectively. Classification accuracy for samples within these subsets reached 89.8% and 91.8%. In cancer data analysis, the gene count was reduced from 60,660 to 10,422, with 3,955 and 6,467 genes in the first and second clusters, respectively. The classification accuracy for these subsets was 97.4% and 97.6%, respectively. To our mind, the implementation of this model promises to significantly improve the efficacy of early diagnosis systems for complex diseases.

The problems of gene regulatory network (GRN) reconstruction and the creation of disease diagnostic effective systems based on genes expression data are some of the current directions of modern bioinformatics. In this manuscript, we present the results of the research focused on the evaluation of the effectiveness of the most used metrics to estimate the gene expression profiles’ proximity, which can be used to extract the groups of informative gene expression profiles while taking into account the states of the investigated samples. Symmetry is very important in the field of both genes’ and/or proteins’ interaction since it undergirds essentially all interactions between molecular components in the GRN and extraction of gene expression profiles, which allows us to identify how the investigated biological objects (disease, state of patients, etc.) contribute to the further reconstruction of GRN in terms of both the symmetry and understanding the mechanism of molecular element interaction in a biological organism. Within the framework of our research, we have investigated the following metrics: Mutual information maximization (MIM) using various methods of Shannon entropy calculation, Pearson’s ${\chi }^2$ test and correlation distance. The accuracy of the investigated samples classification was used as the main quality criterion to evaluate the appropriate metric effectiveness. The random forest classifier (RF) was used during the simulation process. The research results have shown that results of the use of various methods of Shannon entropy within the framework of the MIM metric disagree with each other. As a result, we have proposed the modified mutual information maximization (MMIM) proximity metric based on the joint use of various methods of Shannon entropy calculation and the Harrington desirability function. The results of the simulation have also shown that the correlation proximity metric is less effective in comparison to both the MMIM metric and Pearson’s ${\chi }^2$ test. Finally, we propose the hybrid proximity metric (HPM) that considers both the MMIM metric and Pearson’s ${\chi }^2$ test. The proposed metric was investigated within the framework of one-cluster structure effectiveness evaluation. To our mind, the main benefit of the proposed HPM is in increasing the objectivity of mutually similar gene expression profiles extraction due to the joint use of the various effective proximity metrics that can contradict with each other when they are used alone.

The growing volume and complexity of gene expression data necessitate biologically meaningful and statistically robust methods for feature selection to enhance the effectiveness of disease diagnosis systems. The present study addresses this challenge by proposing a pipeline that integrates RNA-seq data preprocessing, differential gene expression analysis, Gene Ontology (GO) enrichment, and ensemble-based machine learning. The pipeline employs the non-parametric Kruskal–Wallis test to identify differentially expressed genes, followed by dual enrichment analysis using both Fisher’s exact test and the Kolmogorov–Smirnov test across three GO categories: Biological Process (BP), Molecular Function (MF), and Cellular Component (CC). Genes associated with GO terms found significant by both tests were used to construct multiple gene subsets, including subsets based on individual categories, their union, and their intersection. Classification experiments using a random forest model, validated via 5-fold cross-validation, demonstrated that gene subsets derived from the CC category and the union of all categories achieved the highest accuracy and weighted F1-scores, exceeding 0.97 across 14 cancer types. In contrast, subsets derived from BP, MF, and especially their intersection exhibited lower performance. These results confirm the discriminative power of spatially localized gene annotations and underscore the value of integrating statistical and functional information into gene selection. The proposed approach improves the reliability of biomarker discovery and supports downstream analyses such as clustering and biclustering, providing a strong foundation for developing precise diagnostic tools in personalized medicine.