TRANSCRIPTIONAL SIGNATURE FOR TUBERCULOSIS PREDICTION USING ENSEMBLE LEARNING TECHNIQUE

Tuberculosis has the most considerable death rate among diseases caused by a single micro-organism type. The disease is a significant issue for most third world countries due to poor diagnosis and treatment potentials. Early diagnosis of tuberculosis is the most effective way of managing the disease in patients and reducing the mortality rate caused. Despite the several methods that exist in diagnosing tuberculosis, the limitations ranging from the cost in carrying out the test to the time taken to obtain the results have hindered early diagnosis of the disease. There is, therefore, the need to research alternative diagnostic methods that can aid diagnosis. Better testing with non-sputum samples, like blood, is now desirable and sustainable for diagnoses. Currently, blood transcriptional signatures (genes) are being considered since blood is easily accessible and can tell the state of the body at any point in time, and results can be gotten promptly. Lots of research will discover relevant transcriptional signatures to aid tuberculosis detection. These signatures can be easily observed from analyzing blood samples to know genes triggered in the body after tuberculosis-causing bacteria have infected it. This project work aims to develop a predictive model that would help in the diagnosis of TB and also identify relevant signatures that are affiliated with tuberculosis. The method used to carry out this research involved analyzing tuberculosis gene expression data obtained from GEO (Transcript Expression Omnibus) database and identifying relevant genes used to develop a classification model to aid tuberculosis diagnosis. A classifier combination of K.Nearest Neighbor, Bayes, and Support Vector Machine was used to develop the classification model. The weighted voting ensemble technique was used to improve the classification model's performance. The transcriptional signature obtained from the research includes "C4orf41", "GNPAT", "DHX15", "AGGF1", "ANKRD17", "TM2D1", "VAMP4". while the performance accuracy of the ensemble classifier was 0.95 which showed a better performance than the single classifiers which had 0.94, 0.92 and 0.87 obtained from KNN, SVM and NB respectively. The research clearly shows that the identified signatures can help in the early diagnosis of tuberculosis. The developed model can also assist health practitioners in the timely diagnosis of tuberculosis, which would reduce the mortality rate caused by the disease, especially in developing countries