Plant DNA alone can predict hidden seed bacteria
Herbinger J, Ramakrishnan DK, Reißfelder J, Babor M, Höhne M
Seed Saving
The seeds you save from your tomatoes or heirloom beans carry an invisible bacterial community that helps seedlings fight disease and stress, and this research points toward a day when gardeners and breeders could predict that community from a plant's family tree alone.
Every seed carries its own tiny community of bacteria, and scientists wanted to know if you could predict which bacteria live in a seed just by knowing how closely related the plant is to other plants already studied. Using a simple genetic barcode and a new machine-learning approach, they found the answer is often yes, especially for plants with lots of close cousins already sampled, like cabbages or grasses. For plants with few known relatives, the predictions get shakier, showing that family trees only tell part of the story.
Key Findings
A model called HD-KNN, which uses genetic distance between plant species, predicted seed bacterial communities across 61 plant species with an average Jensen-Shannon divergence of 0.276 (lower is better).
Prediction accuracy was highest for well-sampled plant families like Brassicaceae (cabbages/mustards) and Poaceae (grasses), where many close relatives were already in the dataset.
A different model, HD-GPR, worked slightly better for plants with few close relatives in the dataset, showing that the best approach depends on how well-represented a species' family is.
chevron_right Technical Summary
Scientists built a machine-learning tool that predicts which bacteria live inside a plant's seeds just by reading a short DNA marker from the plant itself, no soil or microbe sampling needed. It works best when close plant relatives have already been studied.
Abstract Preview
Original paper
Predicting the seed microbiome using phylogeny-driven machine learning.
The composition of the seed-associated bacterial microbiome can reflect host evolutionary relationships, a pattern consistent with phylosymbiosis. While machine learning offers new opportunities to...
open_in_new Read full abstractAbstract copyright held by the original publisher.
Species Mentioned
Was this useful?
Want to tell us more? (optional)
Thanks for the note!
Something went wrong — please try again.
Too many submissions. Try again in an hour.
Gene editing removes 97% of celiac-triggering proteins from bread wheat
It could mean that people with celiac disease — roughly 1 in 100 worldwide — may one day safely eat bread made from real wheat, without sacrificing the taste...
Brassicaceae or Cruciferae is a medium-sized and economically important family of flowering plants commonly known as the mustards, the crucifers, or the cabbage family. Most are herbaceous plants, while some are shrubs. The leaves are simple, lack stipules, and appear alternately on stems or in r...