Contact us info@computomics.com +49 7071 568 3995
corn header

Innovative Plant Breeding

The machine learning technology behind all future crops.

Innovative plant breeding with xSeedScore, our unique machine learning-based technology.
View on YouTube

 

Accelerate Your Breeding Programs

Computomics' machine learning not only answers your toughest plant breeding questions, but empowers you to ask questions you never had the audacity to pose. Computomics’ SeedScore, Genotyping, and DataScore Technologies provide you with the answers, putting you in control of your plant breeding:

  • Identify 10x more candidates for your commercial pipeline
  • Learn genetic potential of all possible crosses
  • Predict best performers, both for today and future climates
  • Reduce time to market by 3-6 years
  • Accelerate your breeding programs dramatically

Whether your plant breeding goals are commercial, consumer or environmental, Computomics’ team can help move your program way beyond current limitations of hybrid performance prediction, available land for testing and complex trait planning.

Get the answers you need to accelerate your plant breeding program, today and tomorrow.

We are happy to support you on your way to a new commercial product by providing customized technical support adjusted to your specific plant breeding needs.


Predict Better Hybrids with Groundbreaking Machine Learning

Benefit from our machine learning-based regularized kernel methods to predict phenotypes from genome-wide markers. These methods model heterosis and genetic gain. We store the trained predictors to reproducibly analyze next season’s data to make results directly comparable.

 

Correlation of xSeedScore Predicted Phenotype versus Actual Phenotype observed in the field.
Prediction accuracy of a phenotype per location and year comparing: measured phenotype (Target) to predicted phenotype (Prediction) including value distribution.
  • More hybrids to test: Simulate millions of crosses to identify the best ones
  • Better hybrids from the start: Predict how each cross will perform before field-testing.
  • Faster learning cycle: Receive machine learning-based breeding values and phenotypes within 48 hours and improve the model with each cycle.

Fully Understand Your Genotyping Data

We work with you — from desired trait to improved plant. In cooperation with experts in the field of plant gene editing, we offer a full range of services to deliver your desired crop variety. Our comprehensive consultancy applies optimal technologies for your individual goals. We leverage machine learning-based genome analysis to identify the best genome editing targets, then employ a high-quality genome editing service accompanied by extensive quality controls to ensure optimal results. You will gain plant varieties, optimized with your traits of interest, to advance into the commercial pipeline and/or introduce into your breeding programs.

Gain an Unbiased View of Your Genotyping Data

We help you to identify genetic markers for your traits of interest from sequencing data, including single nucleotide polymorphisms (SNPs), insertions and deletions (InDels), copy number variations (CNVs) or structural variants (SVs). By relying on sequencing-based genotyping we ensure an unbiased view of the variance of a population, as it does not rely on previous knowledge.


Depending on your project and specific needs, we set up a tailor-made sequencing-based genotyping pipeline which will take into account optimal sequencing technologies, parameters tuned to arrive at the marker resolution required for your goals and the possibility to impute missing data.

Visualization of identified markers across chromosomes
Visualization of identified markers across chromosomes. Markers are evenly distributed over the chromosomes, which ensures that all important areas of the genome are covered.
Maximize your Genotyping Power by Building a Genome Graph
building the genome graph
Pangenome graphs that incorporate variation from long read data can be utilized to genotype large and complex variants as well as non-reference regions with short read data.

To further increase the accuracy of variant calling, especially for short read data, we can build genome graphs from one or many of these data sources: long read data, assembled contigs or whole genomes, or databases of (structural) variation from your species.

 

This allows you to:

1. Discover reliable SV markers, even for short read or low-coverage data

2. Reduce or even eliminate reference bias in calling variants

3. Gain access to variation in non-reference genome space

 

Accuracy will continuously increase with every newly sequenced genome that is incorporated into the pangenome graph. We also offer the advantage of genotyping your (long forgotten) historic material or incorporate it into the graph to collect all the information about your population in one data structure.

Identify Precisely Novel Genome Editing Targets Through Machine Learning

In addition to traditional plant breeding, new breeding techniques such as CRISPR offer novel possibilities to optimize and improve plant traits. Editing enables targeted, flexible and rapid engineering of plant genomes.

 

Despite the fact that these breakthrough technologies are paving the way to design plants for the future, there are still numerous challenges to overcome. One of them is precise detection and identification of the relevant editing targets that can make a positive impact on a desirable trait. This can be challenging, time-consuming, and labor-intensive, especially for complex traits like yield or diseases resistance.

 

We work with you – from desired impact to improved plant: In partnership with Hudson River Biotechnology (HRB), we offer the most comprehensive end-to-end-solution for your breeding program, to develop varieties that are more resilient, nutritious and capable of delivering great yield even in a changing climate.

Accelerate your plant breeding with our joint solution AccelATrait™ and benefit from the combined experience of experts in bioinformatics, plant breeding and genome editing. Using the latest machine learning and CRISPR genome editing technologies, we bring new plant varieties to market 3-4 times faster than traditional breeding.

 

Benefits of Computomics x HRB AccelATrait™ solution:

  • Detection, identification and location of novel editing targets using the latest machine learning approaches by Computomics
  • Speed up commercialization of genetically uniform edited plants with HRB’s validated CRISPR workflow

 

End-to-end workflow:


Downloads

xSeedScore PDF for download

The machine learning technology behind all future crops

xSeedScore Information Download


Success Story

Predict virtual hybrids from a male and female double-haploid population and predict hybrid phenotypes that exceed their parents and testers

Success Story

Multi-trait optimization in malting barley for specific climates

Success Story

Advancing rice breeding by predicting actual phenotypic values in specific environments


Get in touch with us