Precision Mining of Gene-Disease Associations via Frequent Itemset Analysis and Bioinformatics Integration

Abstract

Biomedical text mining involves the extraction of relevant information from biomedical datasets. It plays a crucial role in genetic research, especially in the development of new drugs where understanding the relationships between genes and diseases is vital. This study introduces a method for generating sets of candidate genes associated with diseases, employing frequent itemset mining for analysis. Genes are ranked based on parameters such as maximum frequent itemset size and gene symbol frequency. This approach aims for precision and efficiency compared to traditional laboratory-based methods, providing highly accurate associations and uncovering novel relationships. Unlike time-consuming laboratory methods, our proposed approach leverages data from the NCBI (National Centre for Biotechnology Information database )via Entrez and utilizes bioinformatics tools like blast for indirect gene associations. Genes exhibiting single nucleotide polymorphisms are identified as indirect genes. The outcomes of this research are anticipated to contribute significantly to biomedical research by offering precise and valuable associations, thereby advancing our understanding of gene-disease relationships..