super.complex_website_humap

Interactive visualizations for protein complexes learned from hu.MAP 1.0 and hu.MAP 2.0 by Super.Complex, along with CORUM complexes are available here. Super. Complex (short for Supervised Complex) is a computational pipeline to predict protein complexes from protein interaction networks making use of a supervised machine learning strategy. Instructions: To view the interactive results, after clicking on the below links, please again click on the different hyperlinks provided on each page. To interact with the visualizations, you can drag or click on the nodes of the graphs of the protein complexes or SARS-COV-2 map.

Interactive visualizations of results on hu.MAP 1.0:

The results below are also available and described in more detail here: https://sites.google.com/view/supercomplex/super-complex-v3-0

Super.Complex predictions:

The results presented below are from [1]:

All Learned human protein complexes, including predicted complexes linked to COVID-19. The complexes here are sorted in descending order of the probability of being a complex (score of the Super.Complex machine learning model).

All Learned Complexes, sortable by protein annotation scores By default, the results are ranked by uncharacterized complex score ( Last column - Candidate score for annotated uncharacterized proteins, computed as the normalized product of the number of proteins with high annotation scores and number of proteins with low annotation scores). The second last column (characterization/annotation score of complex) is the weighted sum of the annotation scores of the proteins.

All Learned Complexes, sorted by number of SARS-CoV-2 interacting hu.MAP 1.0 proteins

SARS-CoV-2 interacting hu.MAP 1.0 protein complexes, sortable by protein annotation scores

hu.MAP 1.0 proteins and their corresponding learned protein complexes

We also sorted hu.MAP 1.0 proteins by their annotation scores, and indicate the learned complexes each protein is a member of, here: hu.MAP 1.0 proteins and their corresponding protein complexes, sorted in ascending annotation score order

We also make available a list of protein complexes that contain proteins interacting with SARS-CoV-2 proteins here: SARS-CoV-2 interacting hu.MAP 1.0 proteins and their corresponding protein complexes

Similarly, sorted in ascending order of annotation score here: SARS-CoV-2 interacting hu.MAP 1.0 proteins and their corresponding protein complexes, sorted in ascending annotation score order

SARS-COV2 Map with predicted complexes from hu.MAP 1.0

CORUM (pre-processed)

All Complexes

All Complexes, sorted by number of SARS-CoV-2 interacting hu.MAP 1.0 proteins

hu.MAP 1.0 proteins and their corresponding protein complexes

CORUM (all)

All Complexes

All Complexes, sorted by number of SARS-CoV-2 interacting hu.MAP 1.0 proteins

hu.MAP 1.0 proteins and their corresponding protein complexes

Interactive visualizations of results on hu.MAP 2.0:

The results presented below are from [2]:

Super.Complex

All Learned Complexes

All Learned Complexes, sortable by protein annotation scores

All Learned Complexes, sorted by number of SARS-CoV-2 interacting hu.MAP 2.0 proteins

SARS-CoV-2 interacting hu.MAP 2.0 protein complexes, sortable by protein annotation scores

hu.MAP 2.0 proteins and their corresponding learned protein complexes

We also sorted hu.MAP 2.0 proteins by their annotation scores, and indicate the learned complexes each protein is a member of, here: hu.MAP 2.0 proteins and their corresponding protein complexes, sorted in ascending annotation score order

We also make available a list of protein complexes that contain proteins interacting with SARS-CoV-2 proteins here: SARS-CoV-2 interacting hu.MAP 2.0 proteins and their corresponding protein complexes

Similarly, sorted in ascending order of annotation score here: SARS-CoV-2 interacting hu.MAP 2.0 proteins and their corresponding protein complexes, sorted in ascending annotation score order

SARS-COV2 Map with predicted complexes from hu.MAP 2.0

CORUM (pre-processed)

All Complexes

All Complexes, sorted by number of SARS-CoV-2 interacting hu.MAP 2.0 proteins

hu.MAP 2.0 proteins and their corresponding protein complexes

CORUM (all)

All Complexes

All Complexes, sorted by number of SARS-CoV-2 interacting hu.MAP 2.0 proteins

hu.MAP 2.0 proteins and their corresponding protein complexes

References

[1] M. V. Palukuri and E. M. Marcotte, “Super.Complex: A supervised machine learning pipeline for molecular complex detection in protein-interaction networks,” PLOS ONE, vol. 16, no. 12, p. e0262056, Dec. 2021, doi: 10.1371/journal.pone.0262056.

[2] M. V. Palukuri, R. S. Patil, and E. M. Marcotte, “Molecular complex detection in protein interaction networks through reinforcement learning.” bioRxiv, p. 2022.06.20.496772. doi: 10.1101/2022.06.20.496772. (Supplement - Table S3, Row 5)

marcottelab / super.complex_website_humap Goto Github PK