Files
Abstract
Bioinformatics is an integral part of systems biology studies, yet many large scale multi-omic studies fail to produce meaningful, actionable results. These experiments produce data sets that are often massive, contain many different types of data in a variety of formats and are often analyzed with a specialized set of tools. Also, scientific and clinical studies often incorporate data sets that cross multiple spatial and temporal scales to describe a particular phenomenon. In this work, these challenges are addressed though the development of a novel analytical framework, Scientific Knowledge and Extraction from Data (SKED), incorporating standardized quantitative data formats, called data primitives, and an extensible object-oriented schema to manage analysis steps. The SKED framework was used to manage analysis of diverse data types from different hosts (three species of non-human primates) and tissues (whole blood, bone marrow, and blood plasma) to investigate molecular mechanisms and interventions to promote host resilience to textit{Plasmodium} infections. Molecular targets that may influence the host response, as well as FDA-approved modulators of these targets, were identified using information from the Pharos database (from the Illuminating the Druggable Genome project) and the Drug-Genome Interaction database. One of these modulators, imatinib, is known to have multiple targets, which were also found here, and the evidence supporting the re-purposing of this drug to promote a resilient host response is presented. This work shows that the SKED approach is able to produce biologically meaningful and verifiable results. The SKED framework is flexible and can be easily extended in the future to new data types, new analysis methods, and other experimental systems.