Why is the study of Lipids important in Data Science?

 

Coupling Machine Learning and Lipidomics as a Tool to Investigate Metabolic Dysfunction-Associated Fatty Liver Diseases

Deep knowledge of disease pathogenesis facilitates diagnosis even with a trivial amount of data and improves the design of clinical research to discover the appropriate cure. Underdiagnosis and undefined clinical predictions characterize nonalcoholic fatty liver disease (NAFLD).

Lipids are among the most important biomolecules, as major components of membranes with crucial tasks in energy distribution and the regulation of extra- and intracellular signaling processes.

Dysregulation of lipid composition in hepatocytes accumulate toxic lipids, and lipotoxicity contributes to the mitochondrial dysfunction, inflammation and deficient liver regeneration that are commonly found in NAFLD.

Most challenges inherent to the analytical, technology-driven discipline of lipidomics may be circumvented, but difficulties in research and the tools for proper management must be identified.

Lipids are prone to oxidation and hydrolysis. Immediate sample processing should be considered, but in clinical research, the procedure is generally not feasible. 

Datasets are acquired over days or weeks, increasing the likelihood for multiple batches and systematic errors.

 
Several software tools have been developed to aid in standardizing and automating the removal of systematic error that work in most lipidomics workflows.

 
In lipidomics workflow, the management of a huge amount of data and their understanding is also a challenging task, which includes the acquisition of metadata.

 There is no laboratory information management system specifically tailored for lipidomics, and they remain partially adapted from those designed for metabolomics, but web-based methods are available, especially in an environment of R programming languages.

Some workflow management systems have been designed as platforms with infrastructure to provide data analysis and algorithms that are appropriated for metabolomics.

Source: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5491763/