Xin Zhou, Ph.D., and Jinghui Zhang, Ph.D., chair, both of St. Jude Computational Biology, helped develop a tool to provide researchers the ability to capture and analyze connections in whole genome data.
As the variety and volume of genomic information grows, so does the challenge of finding lifesaving secrets buried in gigabytes of data.
Computational biologists at St. Jude Children’s Research Hospital have developed a web-based visualization tool called GenomePaint to make the process easier. The tool, including case reports that demonstrate its features, is detailed today in the journal Cancer Cell.
GenomePaint helps researchers capture and analyze connections in whole genome, whole exome, transcriptome and epigenomic tumor sequencing data. Users can visualize and integrate different types of genomic data from individual tumors, tumor subtypes or patient cohorts. The goal is to better understand at the molecular level how cancer begins, spreads and resists treatment. The answers will aid efforts to develop more effective treatments, including precision medicines.
“This tool puts patients’ cancer genomes front and center, helping to recognize patterns that are easy to overlook in rare cancer subtypes or mutations that appear in just a small subset of tumors,” said co-corresponding author Jinghui Zhang, Ph.D., chair of the St. Jude Department of Computational Biology. Rare subtypes account for more than half of childhood cancer.
GenomePaint: Features
GenomePaint can be used to explore thousands of pediatric and adult tumor samples available on St. Jude Cloud. Additional samples will be added regularly.
The tool makes detailed analysis of gene alterations such as structural variations easier. That is also the case for DNA variations that do not encode proteins. These non-coding regions make up about 98% of the human genome, but the impact of such variations has been difficult to analyze with existing web applications.
Other features let users study the functional impact of genomic variations using data generated from cancer cell lines.
GenomePaint is available at no cost to researchers worldwide, who can use the tool to study their own datasets. “The interactive design lets researchers with no computing background carry out in-depth online analysis,” said Xin Zhou, Ph.D., of St. Jude Computational Biology. He is the study’s first and co-corresponding author.
Three interactive and interconnected ways to explore data
GenomePaint lets users “paint” genes or DNA regions of interest and then track various genomic data using different views.
- The cohort view lets researchers “paint” and track DNA variants and RNA expression data in patients with the same tumor subtype. Users can correlate clinical outcomes with DNA genomic alterations or gene expression.
- The matrix view takes a gene-centric approach to understand mutations that drive tumor types. Users can assemble and “paint” genes to evaluate mutational patterns in tumor types as well as associated patient outcomes.
- The sample view offers a unique tumor-centric approach. The view helps researchers identify and integrate the genomic variations driving individual tumors as well as the mechanism involved.
“The sample view is a powerful analytic tool for clinical sequencing data, particularly whole genome sequencing,” Zhang said. The St. Jude clinical genomics program includes an earlier version of GenomePaint, which will be updated with the latest version. Researchers in Australia have also requested a copy of GenomePaint to use in clinical genomic analysis.
The researchers provided examples of how they used GenomePaint to discover new mutations in B-cell acute lymphoblastic leukemia (ALL), the most common childhood leukemia. The new variations were identified in the genes CREBBP and MYC, which are both known cancer-driver genes.
GenomePaint also helped scientists track how mutations in the gene EGFR evolved from diagnosis to relapse in adults with the brain tumor glioblastoma.
Data sources, authors and funding
Study data included 3,854 pediatric cancer genomes published by the St. Jude–Washington University Pediatric Cancer Genome Project and the National Cancer Institute’s TARGET initiative. TARGET is short for Therapeutically Applicable Research to Generate Effective Treatment.
The other authors are Jian Wang, Jaimin Patel, Marc Valentine, Ying Shao, Scott Newman, Edgar Sioson, Liqing Tian, Yu Liu, Samuel Brady, Diane Flasch, Xiaotu Ma, Yanling Liu, Robin Paul, Michael Edmonson, Michael Rusch, Chunliang Li, Suzanne Baker and John Easton, all of St. Jude.
The research was supported in part by grants (CA216391, CA021765) from the National Institutes of Health; and ALSAC, the St. Jude fundraising and awareness organization.
St. Jude Children's Research Hospital
St. Jude Children's Research Hospital is leading the way the world understands, treats and cures childhood cancer and other life-threatening diseases. It is the only National Cancer Institute-designated Comprehensive Cancer Center devoted solely to children. Treatments developed at St. Jude have helped push the overall childhood cancer survival rate from 20% to 80% since the hospital opened more than 50 years ago. St. Jude freely shares the breakthroughs it makes, and every child saved at St. Jude means doctors and scientists worldwide can use that knowledge to save thousands more children. Families never receive a bill from St. Jude for treatment, travel, housing and food — because all a family should worry about is helping their child live. To learn more, visit stjude.org or follow St. Jude on social media at @stjuderesearch.