Predicting autism spectrum disorder from associative genetic markers of phenotypic groups using machine learning

摘要: Machine learning is a discipline of artificial intelligence, geared towards the development various critical applications. Due to its high precision, it widely adopted in process extracting useful hidden patterns and valuable insights from complex data structures. Data extracted real-time environment might contain some irrelevant information. The presence noise degrades model performance. Gene expression an important source, carries genetic information species. pattern reveals significant relationship between genes associated with several diseases. But due irregular molecular interactions reactions occurs during transcription process, gene expressions are minimally affected. It causes detrimental effect on identification biological markers To address this problem, novel selection strategy proposed identify candidate biomarkers genomic data. Signal Noise ratio logistic sigmoid function, Hilbert–Schmidt Independence Criterion Lasso, regularized algorithm amalgamation finds optimal features. system tested microarray dataset autism spectrum disorder (ASD), accessed omnibus repository. FAM104B, CCNDBP1, H1F0, ZER1 identified as ASD. methodical performance evaluation examined used machine algorithms. methodology enhanced prediction rate ASD attained accuracy 97.62%, outperformed existing methods. Also, could act tool assist medical practitioners for accurate diagnosis.