Health AI Central
Peer-Reviewed Publication
Nat Comput Sci2026;6(3):285-300.March 1, 2026Journal Article

Partially shared multi-modal embedding learns holistic representation of cell state.

Xinyi Zhang1,2, G V Shivashankar3,4, Caroline Uhler5,6
1Laboratory for Information and Decision Systems, Massachusetts Institute of Technology, Cambridge, MA, USA.
2Eric and Wendy Schmidt Center, Broad Institute of MIT and Harvard, Cambridge, MA, USA.
3Department of Health Sciences and Technology, ETH Zurich, Zurich, Switzerland. gshivasha@ethz.ch.
4Paul Scherrer Institute, Villigen, Switzerland. gshivasha@ethz.ch.
5Laboratory for Information and Decision Systems, Massachusetts Institute of Technology, Cambridge, MA, USA. cuhler@mit.edu.
6Eric and Wendy Schmidt Center, Broad Institute of MIT and Harvard, Cambridge, MA, USA. cuhler@mit.edu.
PubMed PMC DOI PDF

Abstract

Current technologies enable the simultaneous measurement of diverse data types at the single-cell level. However, data are often processed separately, or integrated via representation learning methods that obscure the contributions of each data modality. Here we present a computational framework that automatically learns partial information sharing between multiple modalities by using an Autoencod…

Create a free account to keep reading

Free members get 10 full research views every month across publications, clinical trials, FDA clearances, adverse events, and NIH grants. No credit card required.

Create Free AccountSign In

Want unlimited research access? See Pro plans

Data Accuracy Notice: Research intelligence on Health AI Central is aggregated from public sources (PubMed, ClinicalTrials.gov, FDA, NIH, CMS, and others) and refreshed nightly. Classifications and derived metrics are produced by automated methods described in our Methodology. We recommend verifying critical data points against the primary sources before making decisions.