Milo2.0 unlocks population genetic analyses of cell state abundance using a count-based mixed model
Kluzer, A.; Marioni, J. C.; Morgan, M. D.
Show abstract
Cell type proportions vary between individuals and are heritable, as demonstrated by statistical genetic analysis of flow cytometry data [1,2]. Higher-resolution cell states can be identified by single-cell RNA-sequencing, the scalability of which now makes it applicable to population-scale cohorts. However, the integration of statistical genetic analysis of cell states using cohort-scale single-cell data requires appropriate algorithms to account for and model the genetic relationships and complex batch-processing inherent to these studies. We describe Milo2.0, which enables the discovery of cell state quantitative trait loci (csQTL), scaling to millions of cells across hundreds of individuals. We identify > 500 csQTLs across peripheral blood immune states and investigate their relationship with the genetic regulation of gene expression. Moreover, we colocalise immune csQTLs with human traits and identify links between immune regulators, cell state abundance and immune-mediated disease.
Matching journals
The top 3 journals account for 50% of the predicted probability mass.
Similar papers in this journal
Similar papers in this journal
- Normalisr: normalization and association testing for single-cell CRISPR screen and co-expression 97%
- Multi-context genetic modeling of transcriptional regulation resolves novel disease loci 97%
- On the discovery of population-specific state transitions from multi-sample multi-condition single-cell RNA sequencing data 96%
Similar papers in this journal
Similar papers in this journal
- Redefining tissue specificity of genetic regulation of gene expression in the presence of allelic heterogeneity 97%
- Factorizing polygenic epistasis improves prediction and uncovers biological pathways in complex traits 96%
- Sparse modeling of interactions enables fast detection of genome-wide epistasis in biobank-scale studies 95%
"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.