Two patients. Same disease. Completely different gene expression data.
Same diagnosis. Same mutation. Same treatment plan.
But Patient X responds to the drug. Patient Y does not.
This happens every day in clinical research. And it confuses data scientists, bioinformaticians, and clinicians alike because if the mutation is the same, the data should look the same. It does not.
Here is why.
All humans carry essentially the same set of genes. But carrying a gene and expressing a gene are two completely different things. The expression of every gene in your body is controlled by a complex web of intrinsic factors like your age, your sex, and your other genes, and extrinsic factors like your environment, your diet, your stress levels, and your history of illness.
This means the same gene, in two different people, under two different conditions, can behave in completely opposite ways.
Now here is where most data scientists make the critical mistake.
They look at Patient X and Patient Y’s gene expression data and ask: why is gene ABCD more highly expressed in Patient X than in Patient Y?
That is the wrong question.
Genes do not work alone. They work in networks. A single gene is almost never responsible for a disease, a drug response, or a clinical outcome. What matters is the pathway, which is the group of genes working together to produce a biological effect.
The right question is: why are genes ABCD, EFGH, and HIJK together more highly expressed in Patient X than in Patient Y? Do they form a pathway? Is something regulating that pathway differently in these two patients?
That shift, from looking at genes individually to looking at genes as networks, is the difference between a data scientist who produces results and a data scientist who produces insights.
And it is not something any AI tool will tell you unprompted. It requires biological understanding.
This is exactly what the Biology Unlocked Journal Club is built for. Every week we take a real paper and ask the right questions together.
Next session: Saturday 27 June. See you all there!