Back

Fast, flexible analysis of differences in cellular composition with crumblr

Hoffman, G. E.; Roussos, P.

2025-01-31 bioinformatics
10.1101/2025.01.29.635498 bioRxiv
Show abstract

Changes in cell type composition play an important role in human health and disease. Recent advances in single-cell technology have enabled the measurement of cell type composition at increasing cell lineage resolution across large cohorts of individuals. Yet this raises new challenges for statistical analysis of these compositional data to identify changes in cell type frequency. We introduce crumblr (DiseaseNeurogenomics.github.io/crumblr), a scalable statistical method for analyzing count ratio data using precision-weighted linear mixed models incorporating random effects for complex study designs. Uniquely, crumblr performs statistical testing at multiple levels of the cell lineage hierarchy using a multivariate approach to increase power over tests of one cell type. In simulations, crumblr increases power compared to existing methods while controlling the false positive rate. We demonstrate the application of crumblr to published single-cell RNA-seq datasets for aging, tuberculosis infection in T cells, bone metastases from prostate cancer, and SARS-CoV-2 infection.

Matching journals

The top 5 journals account for 50% of the predicted probability mass.

1
Nature Communications
○ 5641 papers in training set
Top 6%
22.2%
Similar papers in this journal
2
Communications Biology
○ 993 papers in training set
Top 0.3%
8.8%
Similar papers in this journal
3
Genome Biology
○ 637 papers in training set
Top 1%
7.8%
Similar papers in this journal
4
Nature Methods
○ 385 papers in training set
Top 2%
6.2%
Similar papers in this journal
5
Cell Reports Methods
○ 165 papers in training set
Top 0.2%
5.5%
Similar papers in this journal
50% of probability mass above
6
Cell Systems
○ 201 papers in training set
Top 0.8%
5.1%
7
Cell Genomics
○ 172 papers in training set
Top 1.0%
4.3%
8
eLife
● 5828 papers in training set
Top 28%
4.3%
9
Scientific Reports
○ 3612 papers in training set
Top 33%
3.2%
10
Briefings in Bioinformatics
◐ 354 papers in training set
Top 3%
2.4%
11
Bioinformatics
◐ 1204 papers in training set
Top 6%
2.4%
12
iScience
○ 1154 papers in training set
Top 12%
2.4%
13
PLOS Computational Biology
● 1863 papers in training set
Top 13%
1.9%
14
Genome Medicine
○ 183 papers in training set
Top 3%
1.7%
15
Patterns
○ 78 papers in training set
Top 2%
1.4%
16
Science Advances
● 1243 papers in training set
Top 24%
1.3%
17
Molecular Systems Biology
○ 162 papers in training set
Top 2%
1.3%
18
Cell Reports
○ 1498 papers in training set
Top 23%
1.1%
19
npj Systems Biology and Applications
○ 125 papers in training set
Top 1%
1.1%
20
Nature Biotechnology
○ 172 papers in training set
Top 3%
1.1%
21
The American Journal of Human Genetics
○ 234 papers in training set
Top 2%
1.1%
22
BMC Bioinformatics
○ 457 papers in training set
Top 5%
1.0%
23
Nature Genetics
○ 286 papers in training set
Top 4%
1.0%
24
Human Genetics and Genomics Advances
○ 84 papers in training set
Top 2%
0.8%
25
NAR Genomics and Bioinformatics
◐ 242 papers in training set
Top 4%
0.8%
26
Nucleic Acids Research
◐ 1281 papers in training set
Top 13%
0.8%
27
Proceedings of the National Academy of Sciences
● 2444 papers in training set
Top 45%
0.6%

"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.