Back

Realtime morphological characterization and sorting of unlabeled viable cells using deep learning

Salek, M.; Li, N.; Chou, H.-P.; Saini, K.; Jovic, A.; Jacobs, K. B.; Johnson, C.; Lee, E. J.; Chang, C.; Nguyen, P.; Mei, J.; Pant, K. P.; Wong-Thai, A. Y.; Smith, Q. F.; Huang, S.; Chow, R.; Cruz, J.; Walker, J.; Chan, B.; Musci, T. J.; Ashley, E. A.; Masaeli, M.

2022-03-01 cell biology
10.1101/2022.02.28.482368 bioRxiv
Show abstract

Phenotyping of single cells has dramatically lagged advances in molecular characterization and remains a manual, subjective, and destructive process. We introduce COSMOS, a platform for phenotyping and enrichment of cells based on deep learning interpretation of high-content morphology data in realtime. By training models on an atlas of >1.5 billion images, we demonstrate enrichment of unlabeled cells up to 33,000 fold. We apply COSMOS to multicellular tissue biopsy samples demonstrating that it can identify malignant cells with similar accuracy to molecular approaches while enriching viable cells for functional evaluation. We show high-dimensional embedding vectors of morphology generated by COSMOS without any need for complex sample pre-processing, gating, or bioinformatics capabilities, which enables discovery of cellular phenotypes, and integration of morphology into multi-dimensional analyses. One sentence summaryA novel platform capable of high-throughput imaging and gently sorting cells using deep morphological assessment.

Matching journals

The top 8 journals account for 50% of the predicted probability mass.

1
Nature Communications
○ 5641 papers in training set
Top 15%
12.2%
Similar papers in this journal
2
Cell
○ 431 papers in training set
Top 0.3%
11.6%
Similar papers in this journal
3
Science Advances
● 1243 papers in training set
Top 2%
7.7%
Similar papers in this journal
4
Cell Reports Methods
○ 165 papers in training set
Top 0.2%
5.4%
Similar papers in this journal
5
Nature Biotechnology
○ 172 papers in training set
Top 0.9%
4.7%
6
eLife
● 5828 papers in training set
Top 26%
4.7%
Similar papers in this journal
7
Cytometry Part A
○ 33 papers in training set
Top 0.1%
3.4%
8
Patterns
○ 78 papers in training set
Top 0.6%
3.2%
50% of probability mass above
9
Molecular Systems Biology
○ 162 papers in training set
Top 0.7%
3.1%
10
Genome Biology
○ 637 papers in training set
Top 4%
3.1%
11
iScience
○ 1154 papers in training set
Top 8%
3.1%
12
Communications Biology
○ 993 papers in training set
Top 6%
3.1%
13
Cell Systems
○ 201 papers in training set
Top 2%
2.7%
14
Nucleic Acids Research
◐ 1281 papers in training set
Top 7%
2.3%
15
Scientific Reports
○ 3612 papers in training set
Top 43%
2.3%
16
Proceedings of the National Academy of Sciences
● 2444 papers in training set
Top 29%
1.6%
17
Nature Methods
○ 385 papers in training set
Top 5%
1.6%
18
Cell Reports
○ 1498 papers in training set
Top 24%
1.1%
19
Nature Machine Intelligence
○ 70 papers in training set
Top 2%
1.0%
20
Life Science Alliance
● 285 papers in training set
Top 7%
1.0%
21
Bioinformatics
◐ 1204 papers in training set
Top 8%
1.0%
22
Advanced Science
○ 286 papers in training set
Top 8%
0.9%
23
Cell Reports Medicine
○ 153 papers in training set
Top 5%
0.8%
24
Analytical Chemistry
● 218 papers in training set
Top 2%
0.8%
25
Journal of Cell Biology
● 392 papers in training set
Top 4%
0.8%
26
PLOS Computational Biology
● 1863 papers in training set
Top 20%
0.8%
27
Nature Biomedical Engineering
○ 47 papers in training set
Top 2%
0.8%
28
BMC Methods
○ 15 papers in training set
Top 0.3%
0.6%
29
Lab on a Chip
● 96 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.