Boutros Lab Collaborates on Development of Predictive Cancer Genetics Software
The Boutros Lab has collaborated on the development of a new open-source software to determine the accuracy of computer predictions of genetic variation within tumor samples.
This exciting research is outlined in a new Nature Biotechnology paper entitled ‘A community effort to create standards for evaluating tumor subclonal reconstruction’. Acting as first author, U of T Medical Biophysics PhD candidate Adrianna Salcedo collaborated with a team of researchers from a wide array of institutions, including the University of Toronto, The Francis Crick Institute, UCLA Jonsson Comprehensive Cancer Center, Oregon Health & Science University, the Oxford Big Data Institute and the Ontario Institute for Cancer Research. Dr. Boutros acted as co-lead on the study, alongside Dr. Quaid Morris (U of T), Dr. Peter Van Loo (Francis Crick) and Dr. Kyle Ellrot (OHSU).
Through their collaborative efforts, this diverse team developed and implemented an open-source software that can be used to judge the accuracy of computer predictions and establish a common benchmarking approach to determine the most accurate computational methods for predicting and characterizing genetic diversity within clinical tumor samples. Through the development of a simulation framework, a scoring system was created to determine how accurately each algorithm predicts various measures of genetic diversity, considering factors such as the proportion of cancerous cells in the tumor sample, the number of genetically different groups of cancerous cells in the tumor sample, the proportion of cells within each of these groups, which genetic mutations were in each group, and the genetic relationship between the groups. Through this predictive software, researchers could more accurately model and match real-life tumors, leading to enhanced personalized treatments for cancer patients.