Recent progress in experimental methods has enabled probing of the spatial organization of DNA.Several analyses have indicated that the spatial organization of DNA is non-random and is related to its function.
There is a grand potential for analyzing this data, but the currently available computational tools are lacking in their support.Chromatin 3D data represents a new paradigm in genomics and requires the development of new methods for analysis and interpretation.How this data is represented in computer programs lays the foundation for all further use.Not only does it affect the performance and efficiency of all computations, but it also sets the premises for a programming interface and the ways in which the data can be accessed.
This thesis is an account of the observations made when developing the functionality of a chromatin 3D data analysis for The Genomic HyperBrowser, a web-based tool for genomic computations.Different ways of handling chromatin 3D data are evaluated, with a particular focus on performance and usability.Suggestions and remarks are then made as to how chromatin 3D data can successfully be handled in HyperBrowser specifically, and in computer programs generally.
The effort led to significant performance improvements in a chromatin 3D data analysis performed in HyperBrowser.
In conclusion, performing analyses on the currently available chromatin 3D data is practically feasible through careful design and implementation of data structures and algorithms.However, as experimental methods improve, the increasing size of the data sets will pose new challenges to the computational methods involved.