Development of computational methods for metabolomics
We will develop a computational method to analyze metabolites qualitatively and quantitatively from metabolic data obtained mainly from CE-MS analysis, as well as to perform statistical processing and its biological interpretation in an efficient manner.
In the development of the system, we will be particularly active in machine learning and will utilize metabolome data from the IAB.
We will first learn some basics of metabolomics and existing metabolomic information processing technologies. Then we will implement methods that can reduce various problems of current metabolomic data (e.g., noise in the data, accuracy of peak extraction, accuracy of matching the same peaks among multiple samples, batch effects, reproducibility, integration methods with other omics data, biological interpretation, appropriate computational infrastructure database, etc.) as software.
Participants should have programming experience, but no prior experience in metabolomics is required.