Back

The Language of Cancer: Decoding Cancer Signatures with Language Models and cfRNA Sequencing

Deng, S.; Sha, L.; Jin, Y.; Zhou, T.; Wang, C.; Liu, Q.; Guo, H.; Xiong, C.; Xue, Y.; Li, X.; Li, Y.; Gao, Y.; Hong, M.; Xu, J.; Chen, S.; Wang, P.

2025-03-18 bioinformatics
10.1101/2024.06.29.601341 bioRxiv
Show abstract

We present GeneLLM, a novel large language model that offers a transformative approach to non-invasive cancer detection and biomarker discovery by directly interpreting plasma cell-free RNA (cfRNA) sequences. Unlike traditional annotation-dependent methods, GeneLLM operates without prior knowledge, achieving significantly improved multi-cancer detection accuracy. Critically, GeneLLM identifies novel cfRNAs ( pseudo-biomarkers) originating from previously unannotated genomic regions-overlooked by existing methods-offering new therapeutic targets and insights into intercellular communication. This innovative, cost-effective approach bypasses traditional bioinformatics tools, generating novel pseudo-biomarkers that outperform existing methods even with low-depth sequencing data. Consequently, GeneLLM opens new avenues for biomarker discovery and expands our understanding of the extracellular transcriptomes role in cancer development.

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

Matching journals

The top 10 journals account for 50% of the predicted probability mass.

1
Nature Machine Intelligence
○ 70 papers in training set
Top 0.1%
10.8%
Similar papers in this journal
2
Bioinformatics
◐ 1204 papers in training set
Top 3%
7.7%
Similar papers in this journal
3
Bioinformatics Advances
◐ 203 papers in training set
Top 0.7%
6.6%
Similar papers in this journal
Nature Communications · published here
○ 5641 papers in training set
Top 30%
4.7%
Similar papers in this journal
5
BMC Bioinformatics
○ 457 papers in training set
Top 2%
4.2%
Similar papers in this journal
6
Briefings in Bioinformatics
◐ 354 papers in training set
Top 2%
4.2%
7
Genome Biology
○ 637 papers in training set
Top 3%
3.4%
8
PLOS Computational Biology
● 1863 papers in training set
Top 10%
3.2%
9
Scientific Reports
○ 3612 papers in training set
Top 34%
3.2%
10
Genome Medicine
○ 183 papers in training set
Top 1%
3.2%
50% of probability mass above
11
IEEE Transactions on Computational Biology and Bioinformatics
● 20 papers in training set
Top 0.1%
3.1%
12
Patterns
○ 78 papers in training set
Top 0.7%
2.7%
13
PLOS ONE
● 5266 papers in training set
Top 39%
2.6%
14
GigaScience
◐ 212 papers in training set
Top 1%
2.6%
15
IEEE Journal of Biomedical and Health Informatics
● 37 papers in training set
Top 0.5%
2.3%
16
NAR Genomics and Bioinformatics
◐ 242 papers in training set
Top 2%
2.3%
17
Nucleic Acids Research
◐ 1281 papers in training set
Top 7%
2.3%
18
Computational and Structural Biotechnology Journal
● 242 papers in training set
Top 3%
1.7%
19
BioData Mining
○ 22 papers in training set
Top 0.2%
1.6%
20
Biology Methods and Protocols
◐ 61 papers in training set
Top 0.9%
1.6%
21
BMC Medical Genomics
○ 50 papers in training set
Top 0.7%
1.3%
22
npj Digital Medicine
○ 118 papers in training set
Top 2%
1.3%
23
IEEE/ACM Transactions on Computational Biology and Bioinformatics
● 38 papers in training set
Top 0.7%
1.3%
24
iScience
○ 1154 papers in training set
Top 28%
1.1%
25
Journal of Biomedical Informatics
○ 47 papers in training set
Top 1%
1.0%
26
Genome Research
● 468 papers in training set
Top 6%
1.0%
27
Communications Biology
○ 993 papers in training set
Top 31%
0.8%
28
Cancer Research Communications
● 51 papers in training set
Top 2%
0.8%
29
Computers in Biology and Medicine
○ 128 papers in training set
Top 5%
0.8%
30
JCO Clinical Cancer Informatics
● 22 papers in training set
Top 0.9%
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.