The influx of biomedical measurement technologies continues to define a rapidly changing and growing landscape, multi-modal and uncertain in nature. The focus of the biomedical research community shifted from pure data generation to the development of methodologies for data analytics. Although many researchers continue to focus on approaches developed for analyzing single types of biological data, recent attempts have been made to utilize the availability of multiple heterogeneous data sets that contain various types of data and try to establish tools for data fusion and analysis in many bioinformatics applications. At the heart of this initiative is the attempt to consolidate the domain knowledge and experimental data sources in order to enhance our understanding of highly-specific conditions dependent on sensory data containing inherent error. This challenge refers to granularity: the specificity or mereology of alternate information sources may impact the final data fusion. In an earlier work, we employed data integration methods to analyze biological data obtained from protein interaction networks and gene expression data. We conducted a study to show that potential problems can arise from integrating or fusing data obtained at different granularity levels and highlight the importance of developing advanced data fusing techniques to integrate various types of biological data for analytical purposes. In this work, we explore the impact of granularity from a more formulized approach and show that granularity levels significantly impact the quality of knowledge extracted from the heterogeneous data sets. Further, we extend our previous results to study the relationship between granularity and knowledge extraction across multiple diseases, examining generalizability and estimating the utility of a similar methodology to reflect the impact of granularity levels.