Back

OmicsNavigator: an LLM-driven multi-agent system for autonomous zero-shot biological analysis in spatial omics

LI, Y.; Vakharia, N.; Mayer, A. T.; Luo, R.; Trevino, A. E.; Wu, Z.

2025-07-25 bioinformatics
10.1101/2025.07.21.665821 bioRxiv
Show abstract

Spatial omics provides unprecedented high-resolution insights into molecular tissue compositions but poses significant analytical challenges due to massive data volumes, complex hierarchical spatial structures, and domain-specific interpretive demands. To address these limitations, we introduce OmicsNavigator, an LLM-driven multi-agent system that autonomously distills expert-level biological insights from raw spatial omics data without domain-specific fine-tuning. OmicsNavigator encodes spatial data into concise natural language summaries, enabling zero-shot annotation of structural components, quantitative analysis of pathological relevance, and semantic search of regions of interest using free-form text queries. We evaluated OmicsNavigator on multiple spatial omics studies of kidney cohorts with different phenotypes and biomarker panels, where OmicsNavigator achieved outstanding performances in structural annotations, pathology assessments, and semantic search across studies. OmicsNavigator offers a scalable, interpretable, and modality-agnostic solution for spatial omics analysis.

Matching journals

The top 6 journals account for 50% of the predicted probability mass.

1
Nature Communications
○ 5641 papers in training set
Top 16%
11.6%
Similar papers in this journal
2
Nucleic Acids Research
◐ 1281 papers in training set
Top 2%
10.3%
Similar papers in this journal
3
Bioinformatics
◐ 1204 papers in training set
Top 3%
9.5%
Similar papers in this journal
4
Briefings in Bioinformatics
◐ 354 papers in training set
Top 1%
7.1%
Similar papers in this journal
5
Cell Systems
○ 201 papers in training set
Top 0.6%
6.5%
Similar papers in this journal
6
Genome Biology
○ 637 papers in training set
Top 2%
6.1%
50% of probability mass above
7
Nature Methods
○ 385 papers in training set
Top 2%
5.3%
8
Genome Medicine
○ 183 papers in training set
Top 0.8%
4.3%
9
Patterns
○ 78 papers in training set
Top 0.6%
3.1%
10
Advanced Science
○ 286 papers in training set
Top 2%
3.1%
11
PLOS Computational Biology
● 1863 papers in training set
Top 13%
2.1%
12
Molecular Systems Biology
○ 162 papers in training set
Top 2%
1.7%
13
GigaScience
◐ 212 papers in training set
Top 2%
1.7%
14
Genome Research
● 468 papers in training set
Top 4%
1.7%
15
iScience
○ 1154 papers in training set
Top 20%
1.6%
16
Nature Biotechnology
○ 172 papers in training set
Top 3%
1.6%
17
Genomics, Proteomics & Bioinformatics
188 papers in training set
Top 1%
1.6%
18
Cell Reports Methods
○ 165 papers in training set
Top 2%
1.4%
19
npj Digital Medicine
○ 118 papers in training set
Top 2%
1.4%
20
Bioinformatics Advances
◐ 203 papers in training set
Top 4%
1.0%
21
Nature Machine Intelligence
○ 70 papers in training set
Top 2%
1.0%
22
Proceedings of the National Academy of Sciences
● 2444 papers in training set
Top 38%
1.0%
23
eLife
● 5828 papers in training set
Top 63%
1.0%
24
Small Methods
○ 29 papers in training set
Top 0.8%
0.8%
25
Scientific Reports
○ 3612 papers in training set
Top 74%
0.8%
26
npj Systems Biology and Applications
○ 125 papers in training set
Top 2%
0.8%
27
Cell
○ 431 papers in training set
Top 12%
0.6%
28
Computational and Structural Biotechnology Journal
● 242 papers in training set
Top 8%
0.6%
29
NAR Genomics and Bioinformatics
◐ 242 papers in training set
Top 5%
0.6%
30
Nature Biomedical Engineering
○ 47 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.