Predictive Models for Transcriptome Variations
Exome sequencing (ES) is used to identify the protein-coding regions of genes and is the most advanced standard-of-care genetic test for children born with multiple medical conditions including pediatric brain cancer. However, the rate by which ES is able to diagnose conditions is still low due to inefficiencies in the current process. The current methods used to increase efficiency of ES may however also lead to missed diagnoses of Mendelian disorders, or disorders caused by a genetic abnormality, such as cystic fibrosis, hemophilia, or sickle cell anemia. Researchers in this project will use the comprehensive dataset available through the Pediatric Brain Tumor Atlas to build and test a new algorithm, a set of instructions used by a computer, that could help predict genetic abnormalities, such as transcriptome variations. This algorithm will be widely applicable, enhancing the ability of medical professionals to provide molecular diagnoses for all pediatric patients with suspected Mendelian disorders.
What are the goals of this project?
Researchers will use data on pediatric brain cancers in an effort to test and train new algorithms intended to make Exome sequencing more efficient and accurate.
What is the impact of this project?
This research will be used to develop predictive models that could enhance the ability of medical professionals to diagnose Mendelian disorders that are often missed using currently available analysis and improve patient outcomes.
Why is the CBTN request important to this project?
The data available for analysis through the Pediatric Brain Tumor Atlas will be used to make newly developed predictive models more accurate in regards to patients with pediatric brain cancers.
The Children's Brain Tumor Network contributed to this project by providing access to the Pediatric Brain Tumor Atlas.