Many of the most important discoveries happen by chance. This was the case when researchers discovered that comparing mice lymphoma genome sequences to a reference genome is like comparing apples and oranges, suggesting the same may also be true for human lymphomas.
Instead of using reference genomes, the research team at the University of Colorado Cancer Center recommends that human studies use healthy cells from the same patient as controls when studying genetic changes in cancer.
The study, “Unexpected effects of different genetic backgrounds on identification of genomic rearrangements via whole-genome next generation sequencing,” was published in the journal BMC Genomics.
The research team was exploring how genomic translocations — fragments of DNA that accidentally get cut out and pasted in the wrong location — contribute to B-cell lymphoma in a mouse model when they stumbled over something rather strange.
Their analysis showed that mice with B-cell lymphoma had more than 1,000 translocations, a number so staggering the team assumed they had made a mistake. To rule out any experimental errors, researchers sequenced the genome of three types of normal mice. It turned out these mice had equally as many translocations.
The team went back to compare the genomes of particular mouse strains to that of the reference genome and downloaded new mouse genomic data from the website of Wellcome Trust Sanger Institute in England. The comparison revealed even more translocations.
The team realized that is was not the lymphoma mouse model that was so different, or their method to detect genetic changes that was flawed. There was a real genetic difference between various mouse strains that was far greater than researchers had previously realized.
So when attempting to find translocations that could be driving lymphoma, the team realized that the way to go forward was to compare cancerous cells from a mouse to healthy cells from the same mouse, or at least the same strain.
If this is not done, the many translocations become “noise,” making it difficult for researchers to identify the signal that is responsible for driving cancer. The noise, in this case, are genetic variations that make up natural differences between mouse strains or individuals, and may not affect the likelihood of getting cancer at all.
“Unfortunately, when we have so many events, the artifacts may mask our real events,” Jing Hong Wang, MD, PhD, associate professor at the University of Colorado School of Medicine’s Department of Immunology and Microbiology, and the study’s senior author, said in a news release.
Researchers often make use of reference sequences, as this provides a faster and cheaper way to make comparisons. But this practice may need to be revised.
“Then we started to think about all these human cancer genomic studies,” Wang said. “People use all this sequencing data to show genomic changes in human cancers, but what if these studies have similar comparison problems?”
Wang pointed out that the problem is only relevant when researchers screen for mutations that have not been linked to lymphoma before. When exploring whether a patient carries a known mutation, the method is different and is not affected by other genetic rearrangements.
“People should be their own control,” Wang said. “Instead of working with the published, generic reference genome, we should work with two samples (control vs. cancer) from the same person. Only then can you really figure out what’s going on in your cancer cell genome.”