Data
Inference has ingested and harmonized major publicly available genetic and multi-omic datasets with results from our own studies with globally underrepresented populations, creating the largest single source of human genomic data for drug discovery and development.
Variant Bio has spent years building genomic cohorts from studies with underrepresented populations around the world. These studies feature deep phenotyping across multiple therapeutic areas, whole genome sequencing, and paired multi-omic data. Inference provides access to findings from the data, aligned with our data principles and commitments.
Variant Bio findings unlock discoveries that are impossible in any public dataset. Variants that are rare in European populations may be common in other ancestries, providing the statistical power to uncover novel therapeutic targets.
We've ingested data from major genetic and functional genomic resources, processing hundreds of billions of data points and making hundreds of millions of significant variant-to-phenotype and molecular trait associations instantly queryable. But ingestion is just the start.
Traits are mapped to unified codes across cohorts, enabling true cross-study comparisons and meta-analyses. Multi-level ontologies for ancestry, omics, and analyses can be combined for dynamic filtering.
All summary stats lifted to GRCh38. Effect alleles, betas, and standard errors calculated the same way across studies. Strand flips fixed. Ref/alt alleles verified. You can trust the data.
Genome-wide association studies, rare variant association studies, expression, splicing, protein and metabolite QTLs, differential expression and quantification, all integrated in one platform, all queryable together.
Inference partners can securely upload their own proprietary genetic and multi-omic data. A robust permissions system ensures your data remains private, visible only to your team while gaining the power of our platform tools and harmonized resources.
Analyze your internal data alongside public data. Run cross-study meta-analyses. Let our AI agents discover insights across your combined data. All without ever exposing your proprietary information.
Our dynamic architecture means new data loaded into the platform is available immediately, no waiting for slow data release cycles.
Explore the Platform →