scVIP: personalized modeling of single-cell transcriptomes for developmental and disease phenotypes
Single-cell RNA sequencing reveals cellular heterogeneity, but linking cellular states to individual-level phenotypes remains challenging. We present scVIP, a generative framework that integrates transcriptional profiles and phenotypic markers to learn personalized individual-level embeddings using generative models and cell-type-aware multi-instance learning. scVIP predicts developmental age, disease progression, and neuropathology, while harmonizing datasets with distinct phenotype definitions. The model highlights disease-relevant cell populations and transcriptional programs underlying neurodegeneration.
Authors
- Cindy T. J. van Velthoven (ORCID: https://orcid.org/0000-0001-5120-4546)
- Jeffrey B. Carroll (ORCID: https://orcid.org/0000-0003-1711-8868)
- Kyle J. Travaglini (ORCID: https://orcid.org/0000-0003-3164-6448)
- Qi Qiao (ORCID: https://orcid.org/0000-0003-1240-0506)
- Mariano I. Gabitto (ORCID: https://orcid.org/0000-0001-6911-344X)
- Anamika Agrawal (ORCID: https://orcid.org/0000-0002-1213-2321)
- Shubhabrata Mukherjee (ORCID: https://orcid.org/0000-0003-2522-2884)
- David W. Fardo (ORCID: https://orcid.org/0000-0002-7207-4696)
- Hsin‐Yu Lai
- Andreas Tjaernberg
- Omar Kana
- Ed Lein
- Yehchan Yoo
Institutions
- University of Kentucky (US)
- University of Washington (US)
- Allen Institute (US)
Publication Details
- Journal
- bioRxiv (Cold Spring Harbor Laboratory)
- Published
- 2026-04-22
- DOI
- https://doi.org/10.64898/2026.04.20.717759
- Primary Topic
- Single-cell and spatial transcriptomics
- Type
- article
- Field-Weighted Citation Impact
- 0.00
Funders
- Common Fund