Files and links (1)
Preprint (Author's original)CC BY-NC-ND V4.0, Open Access
Preprint
MultiVI: deep generative model for the integration of multi-modal data
bioRxiv
20/Aug/2021
Abstract
Jointly profiling the transcriptional and chromatin accessibility landscapes of single-cells is a powerful technique to characterize cellular populations. Here we present MultiVI, a probabilistic model to analyze such multiomic data and integrate it with single modality datasets. MultiVI creates a joint representation that accurately reflects both chromatin and transcriptional properties of the cells even when one modality is missing. It also imputes missing data, corrects for batch effects and is available in the scvi-tools framework: https://docs.scvi-tools.org/.
Details
- Title
- MultiVI: deep generative model for the integration of multi-modal data
- Creators
- Tal Ashuach - University of California, BerkeleyMariano Gabitto - Allen InstituteMichael Jordan - University of California, BerkeleyNir Yosef (Corresponding Author) - University of California, Berkeley
- Resource Type
- Preprint
- Publication Details
- bioRxiv
- Number of pages
- 27
- Publisher
- Cold Spring Harbor Laboratory Press; Cold Spring Harbor
- Language
- English
- DOI
- https://doi.org/10.1101/2021.08.20.457057
- Grant note
- We thank Adam Gayoso for assistance on model implementation in scvi-tools. We thank Christina Usher for assistance with visualizations. Author contributions - TA, MIG and NY conceived of the model and designed the analyses. TA and MIG implemented the model with input from AG. TA performed the analyses. NY and MJ supervised the work. TA, MIG, MJ and NY wrote the manuscript.
- Record Identifier
- 993420484203596
Metrics
30 Record Views