The availability of large structured datasets has prompted the need for efficient data analysis and modeling techniques. In systems biology, data-driven modeling approaches create models of complex cellular systems without making assumptions about the underlying mechanisms. In this chapter, we will discuss eigenvalue-based approaches, which identify important characteristics (information) of big datasets through decomposition and dimensionality reduction. We intend to address singular value decomposition (SVD), principle component analysis (PCA), and partial least squares regression (PLSR) approaches for data-driven modeling. In multi-linear systems (that share characteristics such as time points, measurements, etc.), tensor decomposition becomes particularly important for understanding higher-order datasets. Therefore, we will also discuss how to scale up these methods to tensor decomposition using an example dealing with host-cell responses to viral infection.
CITATION STYLE
Shah, M., Chitforoushzadeh, Z., & Janes, K. A. (2016). Statistical Data Analysis and Modeling. In Studies in Mechanobiology, Tissue Engineering and Biomaterials (Vol. 17, pp. 155–175). Springer. https://doi.org/10.1007/978-3-319-21296-8_6
Mendeley helps you to discover research relevant for your work.