Organization of Variation-Based Personal Genetic Data with Document-Based No-Sql Database

Abstract

Variation-based personal genetic data are at the center of many clinical practices and many studies in bioinformatics. Unfortunately, almost all existing methods developed to organize personal genetic data are not variationbased and these methods have not been tested with a large amount of real data. In applications requiring variation-based data, an intense data conversion problem arises when these existing methods are used. On the other hand, the few solutions available that are variation-based are not entirely structural, and they do not meet the needs of daily practice. In this study, a document-based No-SQL database and related designs are proposed for the organization of variation-based personal genetic data. Our structural solution contains many classes, collections and indexes, and it supports all types of variations (both structural and non-structural). In this database, the variation data of 2504 people published by the 1000 Genomes Project were stored smoothly and efficiently. The spaces occupied by personal genetic data in primary memory and hard disk were analyzed. In addition, some queries that might be frequently used by clinical applications were run and the response times of the database was calculated. The results of the analyzes show that the proposed method provides very important gains.