Science1 publisher3 min readPublished
Leaf DNA from abandoned Maine farms fills a predicted blank in the apple pedigree
Cameron Peace's lab at Washington State matched leaves collected from old Maine farms to an apple ancestor his pedigree had inferred but never sampled. The tree that supplied them is still standing.
The Scientist · Science desk
What happened
- Cameron Peace, a tree-fruit geneticist at Washington State University in Pullman, runs MyFruitTree, which takes leaves from unidentified apple trees sent in by the public and matches their genetic fingerprints against a database of known types.
- One sample, coded AMHO-504, matched a predicted empty space high in Peace's North American apple pedigree across its entire DNA fingerprint.
- About 2,500 apple varieties exist in the United States and roughly 100 are grown commercially, but most of the market depends on 15 of them.
Compiled by The ScientistSomething wrong?How this is made
Why it matters
- capability A pedigree that names the ancestors it is missing turns any unlabelled tree into a test: a leaf can be checked against a predicted genotype without documentation to back it up.
- cost The only national figure for fire blight losses is more than two decades old, so what a 15-variety market costs American growers in disease today is not on the public record.
- decision Breeders who want traits that orchardists selected against two or three centuries ago have to obtain them from surviving trees. Nonprofit salvage surveys on abandoned farmland supply that germplasm.
- precedent Peace does these analyses because each one feeds his pedigree, so the search for the next confirmed missing ancestor is driven by the database's own gaps.
A pedigree built from marker data can predict an ancestor nobody has sampled. Peace's continental family tree follows flavor, color, hardiness and disease resistance through siblings, parents and forebears [6], and where the relationships among known varieties cannot be accounted for by any variety in hand, the reconstruction leaves a labelled gap. He ran the Maine sample coded AMHO-504 through a spreadsheet tool that compares a profile against his collection [7]. Its entire fingerprint matched one of those gaps, high in the pedigree [8].
The order of events matters. The blank was predicted from the DNA of varieties already in the database, before the Maine leaves reached Pullman [8]. What the hit establishes is the genotype of the tree standing in Maine, and the ancestry it implies: Scientific American reports that the team had found a living progenitor to hundreds of varieties [9]. Age is a separate question. The magazine calls it possibly the oldest living apple tree in the U.S. [12], on a property that dates to the American Revolutionary War [11].
How often the method works is harder to say. The article does not report how many Maine samples the lab ran, or how many matched anything at all [22], so there is no hit rate to compare against archival provenance. MyFruitTree's public submissions are what keep the reference set growing [2], and Peace keeps doing the analyses because each one feeds the pedigree [6]. This particular leaf came through an organized salvage survey: the Maine Organic Farmers and Gardeners Association was working old and abandoned farms, hoping to rescue rarities before high winds or hard winters killed the trees [4][5].
The market numbers explain why a land-grant lab spends time on orphan trees. About 2,500 apple varieties exist in the U.S. and roughly 100 are grown commercially, but most of the market rests on 15 [15]. Fifteen varieties are about 0.6 percent of the varieties in the country, and they carry most of its sales [20]. Planted that narrowly, orchards give pathogens a stable target: once fire blight, powdery mildew or scab settles on a strategy, it does not need to evolve past the defenses of many different varieties [16]. A U.S.-wide estimate made more than 20 years ago put annual fire blight losses above $100 million [17]. Individual outbreaks since have cost producers between $16 million and $42 million each, which is a per-event regional figure and not a national annual one [18].
The same logic applies to looking backward for traits. "Maybe there are aspects that people didn't value in the past and, 200, 300 years ago, selected against," Peace said. "But these days, actually, we do need them." [13]
What to watch
- Whether Peace's group publishes the marker panel and match statistics for AMHO-504, which would let others judge the false-match probability behind the identification.
- Whether further MOFGA samples from Maine farms match other predicted gaps in the pedigree, giving the method a hit rate instead of a single case.
- Whether any disease-resistance trait from the Maine tree is trialled in an actual breeding program.