English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

The preservation of soil health has been identified as one of the mainchallenges of the XXI century given its vast (and potentially threatening)ramifications in agriculture, human health and biodiversity. Here, we providethe first deep investigation of the predictive potential of machine-learningmodels to understand the connections between soil and biological phenotypes.Indeed, we investigate an integrative framework performing accuratemachine-learning-based prediction of plant phenotypes from biological, chemicaland physical properties of the soil via two models: random forest and Bayesianneural network. We show that prediction is improved, as evidenced by higherweighted F1 scores, when incorporating into the models environmental featureslike soil physicochemical properties and microbial population density inaddition to the microbiome information. Furthermore, by exploring multiple datapreprocessing strategies such as normalization, zero replacement, and dataaugmentation, we confirm that human decisions have a huge impact on thepredictive performance. In particular, we show that the naive total sum scalingnormalization that is commonly used in microbiome research is not the optimalstrategy to maximize predictive power. In addition, we find that accuratelydefined labels are more important than normalization, taxonomic level or modelcharacteristics. That is, if humans are unable to classify the samples andprovide accurate labels, the performance of machine-learning models will belimited. Lastly, we present strategies for domain scientists via a full modelselection decision tree to identify the human choices that maximize theprediction power of the models. Our work is accompanied by open sourcereproducible scripts (https://github.com/solislemuslab/soil-microbiome-nn) formaximum outreach among the microbiome research community.

Human Limits in Machine Learning: Prediction of Plant Phenotypes Using Soil Microbiome Data