Computomics applies machine learning to genomic and environmental data, helping plant breeders predict crop performance under changing climate conditions before multi-year field trials are complete. Its ×SeedScore® platform enables commercial breeding programs to select resilient, high-yielding varieties with greater speed and confidence. The company also offers Pantograph, an interactive omics data hub for pangenome analysis and trait discovery.
Funding
Funding not disclosed
Founders
Product
Problem
Plant breeders face the challenge of developing crop varieties that can withstand rapidly changing climate conditions, yet traditional trial-based selection takes years and cannot predict performance under future environments. This delay leads to varieties entering the market that were selected for a climate that no longer exists, resulting in yield losses and wasted breeding seasons.
Solution
Computomics builds machine learning models that predict how a given genotype will perform under specific environmental conditions by combining genomic data, environmental data such as temperature, rainfall, and soil, and field measurements. Its ×SeedScore® platform runs these predictions at the scale of a commercial breeding program, enabling breeders to answer questions that trial data alone cannot answer in time—such as which candidates hold up under hotter and drier conditions and which are stable across environments. The platform helps breeders identify superior parental combinations, advance candidates before full multi-year data is available, and select trial locations that reflect future markets. This approach accelerates genetic gain and shortens the path to commercialization under changing climate conditions.
Target Audience
Primary customers are commercial plant breeders working in field crops, forages, vegetables, and specialty crops, as well as agri-food companies seeking climate-resilient variety development.
Features
- ×SeedScore® platform for predicting performance across genotype, environment, and management combinations at commercial breeding scale
- Machine learning models that integrate genomic, environmental, and field measurement data to forecast trait performance
- Pantograph, an interactive omics data hub that brings multiple genome assemblies into one shared reference space for pangenome analysis
- Visualization tools that allow users to browse structural variants, genes, and genetic variation from whole-genome down to single base-pair resolution
- Trait discovery capabilities through pangenome analysis, enabling identification of causal genes, novel markers, and gene-editing targets
- Additional tools including CropCompass, BreedScope, MORPHEUS, and MEGAN7 for breeding and microbiome applications