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.

This manuscript explores the application of deep learning (DL) techniques for classifying gene expression data. A key aspect of our research is the comparative analysis of various DL neural network architectures, including Convolution Neural Networks (CNN), Long Short-Term Memory (LSTM) and Gated Recurrent Unit (GRU) Recurrent Neural Networks (RNN), as well as hybrid models that combine these networks. We applied the Bayesian optimization algorithm using 5-fold cross-validation for optimal hyperparameter tuning, which is crucial for DL algorithm performance. Significantly, we have advanced the methods for applying RNNs in processing gene expression data, particularly focusing on LSTM and GRU types. Our study introduces also a novel hybrid quality criterion for data classification, calculated as a weighted sum of partial quality criteria, incorporating an integrated F1-score derived through the Harrington desirability method. Furthermore, we investigate hybrid models that leverage various DL methods, enhancing decision-making objectivity in sample identification. This model uses a step-by-step information processing procedure, initially applying different DL models to gene expression data and subsequently processing these through a CART-based classifier for final decision-making. Our experiments, performed on gene expression data from patients with eight cancer types and one subset with normal samples (without cancer), demonstrated that GRU-RNN-based models, particularly a two-layer GRU-RNN, achieved the highest classification efficacy, with an accuracy of 97.8% on the test dataset. The performance of this model exceeded that of other models, whose accuracy varied between 96.6% and 97.3%. Comparative analysis with other studies in this field suggests that the proposed techniques demonstrate higher efficacy compared to similar research regarding the application of DL models for cancer-type diagnosis.