Back

ColocAI: artificial intelligence approach to quantify co-localization between mass spectrometry images

Ovchinnikova, K.; Rakhlin, A.; Stuart, L.; Nikolenko, S.; Alexandrov, T.

2019-09-08 bioinformatics
10.1101/758425 bioRxiv
Show abstract

MotivationImaging mass spectrometry (imaging MS) is a prominent technique for capturing distributions of molecules in tissue sections. Various computational methods for imaging MS rely on quantifying spatial correlations between ion images, referred to as co-localization. However, no comprehensive evaluation of co-localization measures has ever been performed; this leads to arbitrary choices and hinders method development.\n\nResultsWe present ColocAI, an artificial intelligence approach addressing this gap. With the help of 42 imaging MS experts from 9 labs, we created a gold standard of 2210 pairs of ion images ranked by their co-localization. We evaluated existing co-localization measures and developed novel measures using tf-idf and deep neural networks. The semi-supervised deep learning Pi model and the cosine score applied after median thresholding performed the best (Spearman 0.797 and 0.794 with expert rankings respectively). We illustrate these measures by inferring co-localization properties of 10273 molecules from 3685 public METASPACE datasets.\n\nAvailability and Implementationhttps://github.com/metaspace2020/coloc\n\nContacttheodore.alexandrov@embl.de

Matching journals

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

1
Bioinformatics
◐ 1204 papers in training set
Top 0.4%
34.4%
Similar papers in this journal
2
Bioinformatics Advances
◐ 203 papers in training set
Top 0.2%
9.8%
Similar papers in this journal
3
Briefings in Bioinformatics
◐ 354 papers in training set
Top 1%
6.8%
Similar papers in this journal
50% of probability mass above
4
Frontiers in Bioinformatics
○ 49 papers in training set
Top 0.1%
4.9%
Similar papers in this journal
5
PLOS Computational Biology
● 1863 papers in training set
Top 10%
3.2%
6
Biological Imaging
◐ 15 papers in training set
Top 0.1%
3.2%
Similar papers in this journal
7
Computational and Structural Biotechnology Journal
● 242 papers in training set
Top 2%
2.8%
8
Patterns
○ 78 papers in training set
Top 0.7%
2.7%
9
Journal of Proteome Research
● 234 papers in training set
Top 0.9%
2.4%
10
Scientific Reports
○ 3612 papers in training set
Top 45%
2.1%
11
GigaScience
◐ 212 papers in training set
Top 2%
2.0%
12
Journal of the American Society for Mass Spectrometry
● 37 papers in training set
Top 0.3%
1.7%
13
PLOS ONE
● 5266 papers in training set
Top 52%
1.1%
14
Nature Communications
○ 5641 papers in training set
Top 53%
1.0%
15
Computers in Biology and Medicine
○ 128 papers in training set
Top 4%
1.0%
16
BMC Bioinformatics
○ 457 papers in training set
Top 5%
1.0%
17
Nature Machine Intelligence
○ 70 papers in training set
Top 2%
1.0%
18
NAR Genomics and Bioinformatics
◐ 242 papers in training set
Top 4%
0.9%
19
Cell Reports Methods
○ 165 papers in training set
Top 3%
0.8%
20
Analytical Chemistry
● 218 papers in training set
Top 2%
0.8%
21
Nature Methods
○ 385 papers in training set
Top 6%
0.8%
22
Communications Chemistry
○ 48 papers in training set
Top 2%
0.6%
23
International Journal of Molecular Sciences
○ 494 papers in training set
Top 17%
0.6%
24
iScience
○ 1154 papers in training set
Top 40%
0.6%
25
Metabolites
○ 53 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.