Back

Foundation Model Attributions Reveal Shared Inflammatory Program Across Diseases

Gold, M. P.; Reyes, M.; Diamant, N.; Kuo, T.; Hajiramezanali, E.; Newburger, J. W.; Son, M. B. F.; Lee, P. Y.; Scalia, G.; BenTaieb, A.; Kapadia, S. B.; Tripathi, A.; Corrada Bravo, H.; Heimberg, G.; Biancalani, T.

2025-07-03 bioinformatics
10.1101/2025.06.14.659567 bioRxiv
Show abstract

Determining a genes functional significance within a cellular context has long been a challenge, as absolute expression level is an unreliable indicator. We introduce SIGnature, a framework for scoring gene importance by leveraging attributions derived from single-cell RNA-sequencing (scRNA-seq) foundation models. Attribution scores reduce technical noise, emphasize regulatory genes, and facilitate cross-dataset comparison - a core challenge for scRNA-seq analyses. We developed the SIGnature package as a tool for generating and querying attributions, enabling rapid gene set searches across massive scRNA-seq atlases. We demonstrated its utility using the MS1 monocyte signature, a poorly understood gene program activated in severe COVID-19 and sepsis. Searching 400 studies revealed novel associations between the MS1 signature and multiple hyperinflammatory conditions, including Kawasaki disease. Experimental validation confirmed Kawasaki disease patient serum induces the MS1 phenotype. These findings highlight that SIGnature can uncover shared mechanisms across conditions, demonstrating its power for large-scale signature scoring and cross-disease analysis.

Published in Nature Biotechnology (predicted rank #4) · training set

Matching journals

The top 4 journals account for 50% of the predicted probability mass.

1
Nature Methods
○ 385 papers in training set
Top 0.4%
18.9%
Similar papers in this journal
2
Nature Communications
○ 5641 papers in training set
Top 8%
18.9%
Similar papers in this journal
3
Nature
○ 645 papers in training set
Top 2%
10.0%
Similar papers in this journal
Nature Biotechnology · published here
○ 172 papers in training set
Top 0.3%
8.0%
Similar papers in this journal
50% of probability mass above
5
Genome Biology
○ 637 papers in training set
Top 3%
4.9%
Similar papers in this journal
6
Cell Genomics
○ 172 papers in training set
Top 1%
3.3%
7
Nature Genetics
○ 286 papers in training set
Top 2%
3.2%
8
Nature Biomedical Engineering
○ 47 papers in training set
Top 0.4%
2.8%
9
Nature Machine Intelligence
○ 70 papers in training set
Top 1%
2.2%
10
Cell Systems
○ 201 papers in training set
Top 2%
2.2%
11
Communications Biology
○ 993 papers in training set
Top 14%
1.7%
12
Nature Microbiology
○ 155 papers in training set
Top 2%
1.5%
13
Nature Chemical Biology
○ 119 papers in training set
Top 2%
1.4%
14
Nucleic Acids Research
◐ 1281 papers in training set
Top 10%
1.4%
15
Nature Cell Biology
○ 118 papers in training set
Top 2%
1.2%
16
Proceedings of the National Academy of Sciences
● 2444 papers in training set
Top 37%
1.1%
17
Briefings in Bioinformatics
◐ 354 papers in training set
Top 6%
1.0%
18
eLife
● 5828 papers in training set
Top 64%
0.9%
19
Nature Structural & Molecular Biology
236 papers in training set
Top 3%
0.9%
20
Science
● 477 papers in training set
Top 8%
0.9%
21
Science Advances
● 1243 papers in training set
Top 29%
0.9%
22
Nature Metabolism
○ 69 papers in training set
Top 2%
0.9%
23
Science Immunology
● 88 papers in training set
Top 2%
0.6%
24
Cell
○ 431 papers in training set
Top 11%
0.6%
25
Molecular Systems Biology
○ 162 papers in training set
Top 3%
0.6%
26
Patterns
○ 78 papers in training set
Top 3%
0.6%
27
Nature Computational Science
○ 55 papers in training set
Top 2%
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.