Contact us +49 7071 568 3995
blog plant breeding

Computomics Response to Ivan Baxter's Opinion on Candidate Genes in Plant Biology

We've read the opinion by Ivan Baxter with great interest. Thanks for publishing this opinion on picking candidate genes! Computomics scientists are convinced that data availability and Machine Learning-based analysis will shape the next decade of plant research.

Our proprietary Machine Learning technology is designed to make Phenotype predictions interpretable for unbiased causal gene identification.

We already use AI-based technology to predict plant performance without having to know all causal genes. Observing feature importance allows us to determine genes and environmental factors to obtain manageable candidate lists rapidly.

Machine Learning methods are unique in their ability to track nonlinear, higher-order correlations on levels unattainable by human-invented models. If you want to expedite and increase the accuracy of your predictions, talk to us.


Machine Learning algorithms with high accuracy and enough data can remove bias from your predictions.

FunctionalAnnotation relies heavily on genes annotated and researched beforehand: that’s why GO terms shouldn't be the only resource used. Our annotation solution queries KEGG, Reactome, Pfam and many other databases.

Our Machine Learning-based solution xSeedScore is able to incorporate data from more than 16 different sources into one data-driven prediction, allowing to model GxE, GxG, GxExM interactions, among others.

That's how we are addressing the challenges described in the Baxter opinion already today.

For more information on xSeedScore contact us.

Share on

Get in touch with us