Back

GOAT: Deep learning-enhanced Generalized Organoid Annotation Tool

Bremer, J. P.; Baumdick, M. E.; Knorr, M. S.; Wegner, L. H. M.; Wesche, J.; Jordan-Paiz, A.; Jung, J. M.; Highton, A. J.; Jäger, J.; Hinrichs, O.; Brias, S.; Niersch, J.; Müller, L.; Schreurs, R. R. C. E.; Koyro, T.; Löbl, S.; Mensching, L.; Konczalla, L.; Niehrs, A.; Vondran, F. W. R.; Schramm, C.; Hölzemer, A.; Oldhafer, K. J.; Königs, I.; Kluge, S.; Perez, D.; Reinshagen, K.; Pals, S. T.; Gagliani, N.; Joosten, S. P.; Topf, M.; Altfeld, M.; Bunders, M. J.

2022-09-08 bioinformatics
10.1101/2022.09.06.506648 bioRxiv
Show abstract

Organoids have emerged as a powerful technology to investigate human development, model diseases and for drug discovery. However, analysis tools to rapidly and reproducibly quantify organoid parameters from microscopy images are lacking. We developed a deep-learning based generalized organoid annotation tool (GOAT) using instance segmentation with pixel-level identification of organoids to quantify advanced organoid features. Using a multicentric dataset, including multiple organoid systems (e.g. liver, intestine, tumor, lung), we demonstrate generalization of the tool to annotate a diverse range of organoids generated in different laboratories and high performance in comparison to previously published methods. In sum, GOAT provides fast and unbiased quantification of organoid experiments to accelerate organoid research and facilitates novel high-throughput applications.

Matching journals

The top 3 journals account for 50% of the predicted probability mass.

1
Nature Communications
○ 5641 papers in training set
Top 0.5%
42.4%
Similar papers in this journal
2
Scientific Reports
○ 3612 papers in training set
Top 10%
6.7%
Similar papers in this journal
3
Communications Biology
○ 993 papers in training set
Top 2%
5.5%
Similar papers in this journal
50% of probability mass above
4
Nature Biomedical Engineering
○ 47 papers in training set
Top 0.3%
4.0%
Similar papers in this journal
5
Cell Reports Methods
○ 165 papers in training set
Top 0.4%
4.0%
Similar papers in this journal
6
iScience
○ 1154 papers in training set
Top 7%
3.2%
7
Small Methods
○ 29 papers in training set
Top 0.1%
2.6%
8
Bioinformatics
◐ 1204 papers in training set
Top 6%
2.4%
9
Scientific Data
○ 209 papers in training set
Top 1%
2.1%
10
Molecular Systems Biology
○ 162 papers in training set
Top 1%
1.7%
11
Life Science Alliance
● 285 papers in training set
Top 4%
1.4%
12
eLife
● 5828 papers in training set
Top 56%
1.3%
13
Nature Biotechnology
○ 172 papers in training set
Top 3%
1.3%
14
Disease Models & Mechanisms
139 papers in training set
Top 2%
1.1%
15
Nature Methods
○ 385 papers in training set
Top 6%
1.1%
16
BMC Biology
○ 265 papers in training set
Top 4%
1.0%
17
Autophagy
○ 39 papers in training set
Top 0.5%
0.9%
18
eBioMedicine
○ 183 papers in training set
Top 6%
0.9%
19
Nucleic Acids Research
◐ 1281 papers in training set
Top 14%
0.6%
20
Nature
○ 645 papers in training set
Top 11%
0.6%
21
Advanced Science
○ 286 papers in training set
Top 11%
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.