Structural variants DB : automatization


Structural protein variants analysis requires a huge number of specific indicators to evaluate proteins qualities. Large collections of data are publicly available but many difficulties arise at the time of handling this data-sets: 1) data scattering: fundamental data not available disseminated in different location; 2) data heterogeneity: references may be different from one source to another; 3) missing data: fundamental data are sadly under used and not processed. A strategy to ease variants study process has been elaborated keeping in mind automation, simplicity and extensiveness. This presented tool collection allows a semi-automatic gathering of useful information from different reference sources. After a merging process, missing structural data are computed from normalized retrievals using existing specialized structure analysis software. Finally the whole data-set is formatted to integrate an existing tool whose role is to feed a user friendly visualization tool. The whole development has been tested on a study case : Glucokinase. Many variants of this major enzyme of energetic metabolism have been reported as involved to diseases. Glucokinase enzyme is involved in MODY, a form of diabetes affecting young people. Many studies link disease to structural variants of the enzyme. A preliminary analysis using this tool allowed to sort out a list of interesting variants. An attempt to try to understand underlying key points of the protein structure is then started. Here is presented an easy-to-use, fully integrated and improvable solution allowing efficient variants analysis by providing clear and trustful information variants structural analysis.

Academic publication



NB : modified version 201707, to be considered as new original