Enhancing Cystic Fibrosis Diagnosis through Motif Driven ML Models

The prediction of genetic diseases is a critical area of research in genomics. With the increasing accessibility of DNA sequencing, the identification of genetic variations associated with diseases has become feasible. Cystic fibrosis (CF) is a genetic disorder characterized by the buildup of thick, sticky mucus that can damage various organs in the body. In this study, we proposed a novel approach for classifying CF-affected and not-affected genetic sequences using motif discovery and machine learning techniques. We utilize Gibbs sampling for motif discovery and k-mer representation for sequence transformation. Subsequently, we train and evaluate four machine learning models - Random Forest, Decision Tree, Logistic Regression and Support Vector Machine (SVM) - to classify genetic sequences based on the identified motifs. Our results demonstrate the effectiveness of motif- guided feature selection in improving classification accuracy and highlight the potential of machine learning in CF diagnosis. In our approach, we got more accuracy in the Random Forest model and future researchers can improve the accuracy by collecting more DNA sequences.