Construction of Saudi computational gene models: Applications in healthcare of prevalent genetic disorders

Genetic disorders, particularly autosomal recessive diseases such as sickle cell disease and thalassemia, pose significant public health challenges in Saudi Arabia. High consanguinity rates contribute to their prevalence, necessitating advanced diagnostic approaches. Computational genomics offers innovative solutions for early diagnosis and personalized healthcare. Recent advancements in genomics, such as whole-genome sequencing and CRISPR-Cas9 gene editing, have transformed our understanding of molecular interventions for hereditary diseases. This research focuses on developing computational gene models tailored to the genetic profile of Saudi Arabia, with a particular emphasis on autosomal recessive (AR) diseases. This study aims to develop a hybrid Convolutional Neural Network-Gated Recurrent Unit (CNN-GRU) model specifically designed for diagnosing genetic illnesses prevalent in Saudi Arabia. The model seeks to leverage the strengths of CNNs for spatial feature extraction and GRUs for temporal dependency analysis, providing a robust tool for genomic data interpretation. The proposed CNN-GRU model was applied to a dataset representative of genetic disorders, demonstrating superior performance metrics with an accuracy of 98.8 %, precision of 98 %, recall of 99 %, and an F1-score of 98.5 %. These results surpassed traditional methods such as SVM, RNN, and LSTM-RNN, underscoring the model's efficacy in diagnosing genetic illnesses. Correlation analysis further validated the model's robustness by highlighting the independence of most medical and test parameters across diverse datasets. However, the analysis of the ROC curve identified areas for optimization, suggesting potential improvements to enhance the model's discriminatory power for even more precise diagnostics.