platoseedplatoseed
Multimodal foundation models to predict uncollected patient biology
Strand AI develops foundation models to generate missing bio-data about patients. With this imputed data, pharmaceutical companies can select better patients for their drug trials and shave months from their drug launch timelines. We’ve already trained a multimodal foundation model that integrates spatial biology modalities, beating SOTA at a fraction of the cost.
Strand AI provides multimodal patient data by predicting missing biological modalities from existing samples to support life sciences, biomarker discovery, and drug development. It focuses on enabling researchers to infer gene expression, proteomics, and spatial transcriptomics from routinely collected data.
Strand AI predicts missing biological modalities (e.g., proteomics, transcriptomics, spatial proteomics) from data you already have, enabling researchers to rescue incomplete cohorts, skip expensive assays, and surface predictive biomarkers without re-acquisition. The platform offers models and tooling to impute modalities across cohorts, including specialized capabilities like predicting from H&E slides and genotypes to expand the data landscape for analysis and model training.
Who it’s for: Life sciences teams in pharma/biotech, clinical trial investigators, biomarker discovery groups, and researchers conducting drug discovery and patient stratification.
Company mentions product introductions and access requests, indicating active product development and potential early traction; references to partnerships or customers are not listed in provided text.
Cofounder & CEO at Strand AI ex-Pathos AI, Enable Medicine, Microsoft Research, Element AI. Trained foundation models for biology on the largest patient dataset in existence.
Cofounder & CTO at Strand AI Ex-Enable Medicine Building foundation models to enrich biology data and improve patient outcomes. Oded Falik, CTO, was the Tech Lead for Enable Medicine's spatial biology platform, scaling to 12B+ single-cell annotations and petabyte-scale pipelines. Programming since 8 years old, published first app on iOS app store at 11 years old.
Strand AI builds foundation models to turn sparse clinical trial datasets into complete multimodal profiles.
Strand AI builds foundation models that impute missing multimodal patient data (genomics, imaging, proteomics, etc.) to help pharma and biotech companies select better trial participants and reduce time to launch. They trained a multimodal model predicting spatial proteomics from H&E slides and plan to deploy it at scale for enriching patient data in trials.
Real world training envs for healthcare AI models
AI and Data for Cancer Therapeutics
