Skip to main content

From DNA to Better Tea: The Business Logic of Modern Breeding

· 2 min read
Nelson Lubanga
Nelson Lubanga
Quantitative Geneticist | Plant breeder | Data Scientist

The $15 Question That Could Transform Tea Breeding​

What if spending $15 per seedling could tell you—with 70% accuracy—whether that plant would become a high-yielding bush, years before you invested in field trials?

That’s not science fiction. It’s the reality of genomic selection, and it’s reshaping the economics of crop improvement worldwide. The question is: why isn’t tea keeping up?

The Current System: High Risk, Low Information​

Picture a tea breeder with 2,000 seedlings in a nursery. From these, maybe 500 advance to the next stage. But at that early stage, breeders have almost no reliable information about future performance—only 5% of what they see reflects true genetic potential.
The rest? Environmental noise and random variation.

So they wait.
Preliminary trials (≈3 years). Advanced trials (≈5 years). Elite trials (≈6 years).
After 14–16 years, the best candidates finally emerge. But by then, climate conditions, markets, and pests have all changed.

The Genomic Selection Alternative​

Now imagine sequencing those same 2,000 seedlings at $15 each.
You already have a reference database linking DNA to real-world performance in mature plants. Using that data, you build a prediction model that forecasts how new seedlings will perform—without waiting years.

Suddenly, selection accuracy jumps from 5% to 70%.
You’re not guessing anymore. You’re predicting from the genetic blueprint.

The Economic Transformation​

Simulation studies spanning 40 years of breeding cycles show the impact:

  • Faster cycles: move from 16-year to 2-year breeding cycles
  • Better accuracy: nearly 4× improvement in selection precision
  • Higher returns: 1.65–1.7× more genetic gain at equal or lower cost

So how does adding a $15 genotyping step actually save money?

The Math That Makes It Work​

The savings come from what you no longer need to do:

  • Skip preliminary trials: DNA data replaces 3 years of early testing
  • Smaller trials: better accuracy = fewer candidates to test
  • Faster turnover: 8 breeding cycles in the time it used to take for one

In one cost-constrained simulation, only 800 of 2,000 seedlings were genotyped to stay within budget. Even so, the program achieved 1.7