Back

GRASP: Gene-relation adaptive soft prompt for scalable and generalizable gene network inference with large language models

Feng, Y.; Deng, K.; Guan, Y.

2026-04-14 bioinformatics
10.1101/2025.10.20.683485 bioRxiv
Show abstract

Gene networks (GNs) encode diverse molecular relationships and are central to interpreting cellular function and disease. The heterogeneity of interaction types has led to computational methods specialized for particular network contexts. Large language models (LLMs) offer a unified, language-based formulation of GN inference by leveraging biological knowledge from large-scale text corpora, yet their effectiveness remains sensitive to prompt design. Here, we introduce Gene-Relation Adaptive Soft Prompt (GRASP), a parameter-efficient and trainable framework that conditions inference on each gene pair through only three virtual tokens. Using factorized gene-specific and relation-aware components, GRASP learns to map each pair's biological context into compact soft prompts that combine pair-specific signals with shared interaction patterns. Across diverse GN inference tasks, GRASP consistently outperforms alternative prompting strategies. It also shows a stronger ability to recover unannotated interactions from synthetic negative sets, suggesting its capacity to identify biologically meaningful relationships beyond existing databases. Together, these results establish GRASP as a scalable and generalizable prompting framework for LLM-based GN inference.

Matching journals

The top 7 journals account for 50% of the predicted probability mass.

1
Nature Communications
4913 papers in training set
Top 14%
12.4%
2
Cell Systems
167 papers in training set
Top 1.0%
10.1%
3
Bioinformatics
1061 papers in training set
Top 3%
9.2%
4
Nature Machine Intelligence
61 papers in training set
Top 0.3%
7.2%
5
Nature Biotechnology
147 papers in training set
Top 2%
4.9%
6
Proceedings of the National Academy of Sciences
2130 papers in training set
Top 16%
4.3%
7
Genome Research
409 papers in training set
Top 0.7%
4.3%
50% of probability mass above
8
Nature Methods
336 papers in training set
Top 2%
4.0%
9
Nucleic Acids Research
1128 papers in training set
Top 7%
3.1%
10
Genome Biology
555 papers in training set
Top 3%
2.6%
11
Science
429 papers in training set
Top 11%
2.6%
12
Nature
575 papers in training set
Top 9%
2.4%
13
Advanced Science
249 papers in training set
Top 8%
2.1%
14
Bioinformatics Advances
184 papers in training set
Top 2%
1.9%
15
Genome Medicine
154 papers in training set
Top 4%
1.7%
16
Briefings in Bioinformatics
326 papers in training set
Top 4%
1.7%
17
Nature Computational Science
50 papers in training set
Top 0.6%
1.7%
18
Nature Medicine
117 papers in training set
Top 2%
1.7%
19
PLOS ONE
4510 papers in training set
Top 57%
1.5%
20
PLOS Computational Biology
1633 papers in training set
Top 19%
1.3%
21
Scientific Reports
3102 papers in training set
Top 66%
1.2%
22
Nature Biomedical Engineering
42 papers in training set
Top 1%
1.0%
23
Nature Genetics
240 papers in training set
Top 6%
1.0%
24
Cell Genomics
162 papers in training set
Top 5%
1.0%
25
IEEE Transactions on Computational Biology and Bioinformatics
17 papers in training set
Top 0.5%
0.9%
26
The American Journal of Human Genetics
206 papers in training set
Top 4%
0.7%
27
iScience
1063 papers in training set
Top 35%
0.7%
28
Frontiers in Genetics
197 papers in training set
Top 11%
0.6%
29
Communications Biology
886 papers in training set
Top 29%
0.6%
30
BMC Bioinformatics
383 papers in training set
Top 8%
0.6%