Back

Machine learning general transcriptional predictors of plant disease

Sia, J.; Zhang, W.; Cheng, M.; Bogdan, P.; Cook, D. E.

2023-09-01 plant biology
10.1101/2023.08.30.555529 bioRxiv
Show abstract

Plants utilize an innate immune system to defend against all classes of microbial invaders. While we understand specific genetic determinants of host-pathogen interactions, it remains less clear how generalized the immune response is to diverse pathogens. Using a data-driven approach, and utilizing feature selection based on network science and topology, we developed machine learning models that could predict host disease development across diverse pathosystems. These machine learning models identified early transcriptional responses predictive of later disease development, regardless of pathogen class, using a fraction of the host transcriptome. The identified gene sets were not enriched for canonical defense genes, but where statistically enriched for genes previously identified from independent data sets, including those described as representing a general plant stress response. These results highlight novel components of a general plant immune response, and demonstrate the application of machine learning to address biological hypotheses of a complex multigenic outcome. TeaserA machine learning approach can predict plant disease development caused by diverse microbial invaders, and newly identified genes may represent novel components of a general plant response to infection.

Matching journals

The top 7 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.