Back

MotorBench: A Cryo-Electron Tomography Dataset of Bacterial Flagellar Motors for Testing Detection Algorithms

Owens, C. B.; Morse, B. S.; Darley, A. J.; Hart, T. J.; Webb, R.; Maggi, S.; Jensen, G. J.; Ward, M. M.; Reade, W. C.; Kaplan, M.; Hart, G. L. W.

2025-04-23 molecular biology
10.1101/2025.04.23.650258 bioRxiv
Show abstract

Understanding bacterial nanomachines like flagellar motors, which are crucial for pathogenic bacteria motility, is vital for microbiological and therapeutic research. Cryogenic electron tomography (cryo-ET) enables visualization of these structures within cells at near-native conditions. But manual identification remains challenging due to low contrast, limited resolution, and crowded in vivo environments. To address this, we introduce MotorBench, an expert-annotated dataset of bacterial flagellar motors that has been curated as part of a Kaggle competition BYU - Locating Bacterial Flagellar Motors 2025, engaging data scientists globally to create automated detection algorithms. MotorBench and its accompanying tools are intended to serve as a benchmark for evaluating and comparing future algorithms in automated cryo-ET analysis.

Matching journals

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

1
Nature Methods
○ 385 papers in training set
Top 0.6%
15.2%
Similar papers in this journal
2
Acta Crystallographica Section D Structural Biology
● 59 papers in training set
Top 0.1%
7.9%
Similar papers in this journal
3
IUCrJ
● 32 papers in training set
Top 0.1%
6.8%
Similar papers in this journal
4
Journal of Structural Biology: X
○ 17 papers in training set
Top 0.1%
6.3%
Similar papers in this journal
5
Journal of Structural Biology
○ 64 papers in training set
Top 0.1%
6.3%
Similar papers in this journal
6
Biology Methods and Protocols
◐ 61 papers in training set
Top 0.1%
5.5%
7
PLOS Computational Biology
● 1863 papers in training set
Top 7%
4.9%
50% of probability mass above
8
PLOS ONE
● 5266 papers in training set
Top 35%
3.3%
9
eLife
● 5828 papers in training set
Top 35%
3.2%
10
Optics Express
● 26 papers in training set
Top 0.1%
2.8%
11
Scientific Reports
○ 3612 papers in training set
Top 41%
2.4%
12
BMC Bioinformatics
○ 457 papers in training set
Top 3%
2.4%
13
GigaScience
◐ 212 papers in training set
Top 2%
2.1%
14
Biological Imaging
◐ 15 papers in training set
Top 0.1%
2.1%
15
Bioinformatics
◐ 1204 papers in training set
Top 6%
2.1%
16
Methods
○ 34 papers in training set
Top 0.2%
1.7%
17
Nature Communications
○ 5641 papers in training set
Top 44%
1.7%
18
Computer Methods and Programs in Biomedicine
○ 28 papers in training set
Top 0.6%
1.5%
19
International Journal of Molecular Sciences
○ 494 papers in training set
Top 9%
1.5%
20
npj Imaging
○ 12 papers in training set
Top 0.2%
1.0%
21
Briefings in Bioinformatics
◐ 354 papers in training set
Top 6%
1.0%
22
Protein Science
○ 246 papers in training set
Top 3%
0.9%
23
Communications Biology
○ 993 papers in training set
Top 34%
0.6%
24
Frontiers in Bioinformatics
○ 49 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.