Back

GFMBench-API: A Standardized Interface for Benchmarking Genomic Foundation Models

Larey, A.; Dahan, E.; Amit Bleiweiss, A. B.; Kellerman, R.; Leib, G.; Nayshool, O.; Ofer, D.; Zinger, T.; Dominissini, D.; Rechavi, G.; Bussola, N.; Lee, S.; O'Connell, S.; Hoang, D.; Wirth, M.; W. Charney, A.; Shavit, Y.; Daniel, N.

2026-02-19 genomics
10.64898/2026.02.19.706811 bioRxiv
Show abstract

The rapid scaling of Genomic Foundation Models (GFMs) has created a critical need for standardized evaluation frameworks. Current benchmarking practices are often fragmented, relying on model-specific preprocessing and inconsistent metric implementations that hinder reproducible comparisons. We present GFMBench-API, a high-level Python interface designed to unify the evaluation lifecycle of GFMs. GFMBench-API provides a modular "middleware" architecture that decouples model-specific tokenization and embedding logic from task-specific data streams and performance metrics. By standardizing the input/output schemas for common genomic tasks, such as regulatory element prediction, variant effect scoring, and long-range interaction mapping, GFMBench-API enables researchers to integrate new models or tasks with minimal "glue code." Our interface ensures mathematical consistency across evaluations, providing a robust foundation for the transparent and systematic benchmarking of GFMs.

Matching journals

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

1
Bioinformatics
◐ 1204 papers in training set
Top 0.8%
25.8%
Similar papers in this journal
2
Genome Biology
○ 637 papers in training set
Top 0.9%
11.5%
Similar papers in this journal
3
Nucleic Acids Research
◐ 1281 papers in training set
Top 3%
7.0%
Similar papers in this journal
4
Bioinformatics Advances
◐ 203 papers in training set
Top 0.7%
6.5%
Similar papers in this journal
50% of probability mass above
5
NAR Genomics and Bioinformatics
◐ 242 papers in training set
Top 1.0%
4.2%
Similar papers in this journal
6
Nature Communications
○ 5641 papers in training set
Top 31%
4.2%
7
PLOS Computational Biology
● 1863 papers in training set
Top 9%
3.9%
8
Nature Computational Science
○ 55 papers in training set
Top 0.2%
3.3%
9
Nature Methods
○ 385 papers in training set
Top 3%
3.1%
10
Genome Research
● 468 papers in training set
Top 2%
3.1%
11
BMC Bioinformatics
○ 457 papers in training set
Top 3%
3.1%
12
GigaScience
◐ 212 papers in training set
Top 1%
3.1%
13
Briefings in Bioinformatics
◐ 354 papers in training set
Top 3%
3.1%
14
GENETICS
◐ 483 papers in training set
Top 2%
2.1%
15
Cell Genomics
○ 172 papers in training set
Top 3%
1.6%
16
Genome Medicine
○ 183 papers in training set
Top 4%
1.1%
17
Computational and Structural Biotechnology Journal
● 242 papers in training set
Top 6%
1.0%
18
Nature Biotechnology
○ 172 papers in training set
Top 5%
0.8%
19
The American Journal of Human Genetics
○ 234 papers in training set
Top 3%
0.8%
20
Database
◐ 61 papers in training set
Top 1%
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.