Towards the Understanding of the Human Genome: A Holistic Conceptual Modeling Approach

Understanding the human genome is a great scientific challenge, whose achievement requires effective data manipulation mechanisms. The non-stop evolution of both new knowledge and more efficient sequencing technologies generates a kind of genome data chaos. This chaos complicates the use of computational resources that obtain data and align them into specific actions. Conceptual model-based techniques should play a fundamental role in turning data into actionable knowledge. However, current solutions do not give a crucial role in the task of modeling that it should have to obtain a precise understanding of this domain. Hundreds of different data sources exist, but they have heterogeneous, imprecise, and inconsistent data. It is remarkably hard to have a unified data perspective that covers the genomic data from genome to transcriptome and proteome, which could facilitate semantic data integration. This paper focuses on how to design a conceptual model of the human genome that could be used as the key artifact to share, integrate, and understand the various types of datasets used in the genomic domain. We provide a full conceptual picture of relevant data in genomics and how semantic data integration is much more effective by conceptually integrating the diverse types of existing data. We show how such a conceptual model has been built, focusing on the conceptual problems that were solved to adequately model concepts whose knowledge is under constant evolution. We show how the use of the initial versions of the conceptual model in practice has allowed us to identify new features to incorporate in the model, achieving a continuous improvement process. The current version is ready to be used as the key artifact in projects where conceptually combining multiple levels of data helps to provide valuable insights that would be hard to obtain without it.