# Examples of Applications **Systematic Knowledge Discovery: Literature Aggregation and Text Mining** With an exponentially growing number of scientific publications (global scientific output doubles every nine years ), manual knowledge collection and curation has become an extremely 24 challenging task. Networks of institutions continuously aggregate new knowledge in thousands of knowledge bases using both manual curation and various automated methods. A single experiment produces thousands to millions of distinct measurements that must be sifted through by referencing this existing knowledge to construct a causal hypothesis explaining the phenomena under study. Automating searches of the body of scientific literature and of the experimental findings specific to the user’s research question is a crucial goal in the application of AI to biomedical research. **Systematic Knowledge Discovery** In silico experiments and analyses use mathematical modeling and computer simulations to overcome various limitations of in vivo and in vitro methods and support the needs and research challenges of the biomedical and pharmaceutical industries. The empirical and physics-based in silico models allow preliminary discovery and testing of novel genetic and metabolic networks to be validated in experiments. The reconstruction process for genome-scale metabolic networks is well developed but labor intensive. Thiele and Palsson published the best protocol in this area 25 of research. However, even with the impressive progress in computational biology and chemistry, the number of tissue- and organ-level simulations is limited. So far, only three organs—the mouse pancreas, the C. elegans gonad—and partial rodent brain development—have been modeled in silico. On the other hand, some models—for example, the human body physiology models developed within the *Physiome project* and the *Virtual Physiological Human* initiatives—have already been applied to solve some clinical problems and have brought in silico modeling closer to clinical translation. **Diagnostic Biomarker Discovery** Biomarkers indicate alterations in one’s biological state or health condition. The discovery of novel biomarkers and advances in high-throughput technologies, such as DNA microarrays and mass spectrometry, provide direct support in observational and analytic epidemiology, clinical trials, screening, diagnosis, and prognosis. Many statistical and machine learning methods have been adopted for measurement and evaluation purposes and for building predictive models based on biomedical data. **Drug Target Discovery** One of the major challenges in biomedical sciences is identifying the metabolic and regulatory pathways of disorders for rational drug design and target-oriented drug development. A simulation of a metabolic network in silico allows for simulated testing of these predicted genotype-phenotype-drug metabolic pathways. **In Silico Patient Modeling for Personal/Precision Medicine Diagnosis and Treatment Planning** The future of medicine will be highly personalized, catering holistically to each patient’s unique biological blueprint. Science is beginning to uncover the unique dynamics of each person’s biological structure by using machine learning tools to piece together a full atlas of an individual’s genomics, proteomics, and other “-omics.” Stronger models that connect our individual microbiomes to our genomes, metabolomes, and epigenomes are beginning to uncover the delicate connections that these factors have in an individual’s body. Once we fully understand these connections, we will be able to bridge accurate diagnosis techniques with highly targeted therapy (so-called theranostics), develop successful strategies for creating high-impact therapeutics, and “shift the emphasis in medicine from reaction to prevention and from disease to wellness.

Figure 12. Main focus areas for developing and sustaining a digital patient

--- # Agent Instructions: Querying This Documentation If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question. Perform an HTTP GET request on the current page URL with the `ask` query parameter: ``` GET https://osiris-2.gitbook.io/osiris/high-level-ai-services/ai-services-provided-by-the-osiris-foundation/bio-data-analytics/examples-of-applications.md?ask= ``` The question should be specific, self-contained, and written in natural language. The response will contain a direct answer to the question and relevant excerpts and sources from the documentation. Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.