Don’t trust. Just verify. That could be the motto of the Cancer Dependency Map, a project to pinpoint the genes that are critical for the survival of cancer cells. Such genes have been cataloged in various datasets. But whether the datasets match each other hasn’t always been clear. Encouragingly, two of the largest datasets—both derived from pan-cancer CRISPR-Cas9 genetic screens—have been compared and found to be consistent.
Even better, the CRISPR-Cas9 screens used to generate the two datasets differed methodologically. Neither screen, then, is likely to replicate the biases of the other. Instead, the screens likely verify each other’s results.
One screen comes from the Wellcome Sanger Institute, the other comes from the Broad Institute. Now that the screens have been integrated, they form the basis for the Cancer Dependency Map. According to the Broad Institute, cancer-specific dependencies are compelling therapeutic targets. That is, the genetic mutations that cause cancer cells to grow also confer specific vulnerabilities that normal cells lack.
Unfortunately, for most cancers, the relationships between the genetic features of cancer and cancer dependencies are poorly understood. Hence the need for the Cancer Dependency Map, which is derived from experiments with cancer cell lines in the laboratory. In these experiments, researchers use CRISPR-Cas9 technology to edit the genes in cancer cells, turning them off one-by-one to measure how critical they are for cancer cell survival. The most essential genes, which are known as dependencies, are the most likely to make viable drug targets.
Detailed findings from a comparison of the Wellcome Sanger findings and the Broad findings appeared December 20 in Nature Communications, in an article titled, “Agreement between two large pan-cancer CRISPR-Cas9 gene dependency data sets.”
“Despite significant differences in experimental protocols and reagents, we find that the screen results are highly concordant across multiple metrics with both common and specific dependencies jointly identified across the two studies,” the article’s authors wrote. “Furthermore, robust biomarkers of gene dependency found in one data set are recovered in the other.”
Through further analysis and replication experiments, Wellcome Sanger scientists and Broad Institute scientists showed that batch effects are driven principally by two key experimental parameters: the reagent library and the assay length.
“This is the first analysis of its kind and is really important for the whole cancer research community,” said Aviad Tsherniak, a corresponding author of the current study and the associate director of Cancer Data Science group at the Broad Institute. “Not only have we reproduced common and specific dependencies across the two datasets, but we have taken biomarkers of gene dependency found in one dataset and recovered them in the other. Our analysis has been unbiased, rigorous, and proves the veracity of the approach and the drug targets identified.”
According to a Wellcome Sanger announcement, the current study validates the reproducibility of CRISPR-Cas9 functional genetic screens, alleviates doubts about their efficacy, and sets rigorous standards for assessing pan-cancer CRISPR-Cas9 gene dependency datasets. Overall, the study advances the comparison and integration of large databases of cancer dependencies. It has already resulted in a combined dataset of sufficient scale—incorporating 725 cancer models and spanning 25 different cancer types—to accelerate the discovery and development of new cancer drugs.
The Cancer Dependency Map aims to bridge the translational gap that exists between genomic sequencing and providing precision medicine to the many cancer patients. In other words, the idea behind the Cancer Dependency Map is to create a detailed rulebook of precision cancer treatments for patients.
“It is worth remembering that when these datasets were originally produced, we were dealing with a new, unproven technology,” noted Francesco Iorio, PhD, principal staff scientist at the Wellcome Sanger Institute and a senior bioinformatician at the Open Targets initiative. This study is important because it demonstrates the validity of the experimental methods and the consistency of the data that they produce. It also means that two large cancer dependency datasets are compatible. By joining them together, we will have access to much greater statistical power to narrow down the list of targets for the next generation of cancer treatments.”