Abstract
Background: The introduction of next generation sequencing (NGS) has revolutionized molecular diagnostics, though several challenges remain limiting the widespread adoption of NGS testing into clinical practice. One such difficulty includes the development of a robust bioinformatics pipeline that can handle the volume of data generated by high-throughput sequencing in a cost-effective manner. Analysis of sequencing data typically requires a substantial level of computing power that is often cost-prohibitive to most clinical diagnostics laboratories. Findings. To address this challenge, our institution has developed a Galaxy-based data analysis pipeline which relies on a web-based, cloud-computing infrastructure to process NGS data and identify genetic variants. It provides additional flexibility, needed to control storage costs, resulting in a pipeline that is cost-effective on a per-sample basis. It does not require the usage of EBS disk to run a sample. Conclusions: We demonstrate the validation and feasibility of implementing this bioinformatics pipeline in a molecular diagnostics laboratory. Four samples were analyzed in duplicate pairs and showed 100% concordance in mutations identified. This pipeline is currently being used in the clinic and all identified pathogenic variants confirmed using Sanger sequencing further validating the software.
Original language | English (US) |
---|---|
Article number | 314 |
Journal | BMC Research Notes |
Volume | 7 |
Issue number | 1 |
DOIs | |
State | Published - May 23 2014 |
Bibliographical note
Funding Information:The authors would like to thank Dr. Anne-Francoise Lamblin, the Research Informatics Support Systems (RISS) program director and project lead for Galaxy Bioinformatics workbench, Shawn Houston, manager for Infrastructure Operations and User Support group and Dan Debertin, Galaxy system administrator at the Minnesota Supercomputing Institute (MSI). The authors would also like to thank MSI at large for their support on this project without which this project would not have been possible. This work was generously supported by funds from the Institute of Translational Neuroscience at the University of Minnesota.
Keywords
- Cloud computing
- Molecular diagnostics
- Next generation sequencing
- Variant detection