Mutations carried half.
Expression carried all.
Inside the nose grows a lump called an inverted papilloma. Under the microscope it is not cancer. Yet one to four out of ten become cancer, and the microscope cannot tell you which.
One paper took matched normal mucosa, papilloma and carcinoma from 11 patients and read the exome, the transcriptome and the mitochondrial genome of all three. The three rulers disagreed. Over these 20 lessons you measure all three yourself and decide which one to trust. Python runs right inside this page. There is nothing to install.
How you'll learn
Normal, papilloma and carcinoma are three states, not two. The usual pairwise comparison cannot see a staircase. You verify by hand why a mixed linear model is needed.
Exome, RNA and mitochondrial DNA give three different answers from the same tissue: five of ten, all of them, none at all. The disagreement is not an error โ it is the finding.
The first pass of this analysis found zero EGFR mutations. Removing one filter box brought them back in four patients. Filters do not only remove noise.
In the last two lessons you rebuild the PandaOmics target list under the same rules. Which boxes you switch on decides who comes first.