Cambridge Healthtech Institute’s Ann Nguyen recently interviewed Mohamed Abouelhoda of King Faisal Specialist Hospital and Research Centre. Dr. Abouelhoda shares his presentation on “Multicloud in Action: A System for Variant Detection and Annotation Workflow in Multiple Clouds” at the Cloud Computing conference taking place May 23-25, 2017 at Bio-IT World Conference & Expo in Boston, MA.
LOW-COST COMPUTATION FOR LARGE-SCALE GENOME PROJECTS
Q1: Can you describe your current activities with the Saudi Human Genome Project?
I am heading the IT and bioinformatics team of the Saudi Human Genome Project. I am responsible for the establishment and operation of the IT infrastructure, management of sample information, and analysis and management of genomics data.
Q2: What workflows have proven especially promising in enabling lower-cost computation for large-scale genome projects?
To reduce the operational costs, we decided to use open source technologies and switch to open industry standards for hardware. Our workflows are for identifying variants in the genomes of the patients and evaluating its role in the disease. For variant identification, we use open source mapping and variant calling tools (Torrent Suite for Ion data and BWA-GATK for Illumina data). The annotation of the variants helps in identifying its role in the disease, and to this end we use the open source annovar system based on public knowledge databases as well as in-house developed population databases.
Q3: What challenges must still be overcome to reduce the costs of genomic data analysis while also achieving research goals?
The major spending is on the storage systems, and it is important to switch to open industry standards to come up with low-cost but efficient and fast storage systems. It is also important to improve compression techniques, either during the analysis or for backup and archiving. Reference-based compression techniques dramatically reduce the storage requirements and these should be widely used.
Mohamed Abouelhoda, Ph.D., Head, Bioinformatics, Saudi Human Genome Project, Genetics Department, King Faisal Specialist Hospital and Research Centre (KFSHRC)
Prof. Dr. Mohamed Abouelhoda is an associate professor of bioinformatics and currently heading the bioinformatics team of the Saudi Human Genome Program. Abouelhoda has more than 16 years of experience in bioinformatics since his work in Bielefeld University in 2001. Abouelhoda received his Ph.D. from Ulm University in Germany in 2005 and worked as a postdoc for the other two years in Ulm University. Since 2008, he has been faculty member in Cairo University and Nile University and led many international projects in the fields. He obtained many academic and industrial awards and is a continuous reviewer in many bioinformatics journals and conferences.
Learn more about Dr. Abouelhoda’s presentation during the Cloud Computing conference: www.Bio-ITWorldExpo.com/Cloud-Computing/