SPACE: STRING proteins as complementary embeddings

Note

📢 Update (October 2026): SPACE embeddings are now available for STRING v12.5! Cross-species network embeddings and ProtT5 sequence embeddings for all eukaryotic proteins in STRING v12.5 can be downloaded from the STRING website:

• protein.network.embeddings.v12.5.h5
• protein.sequence.embeddings.v12.5.h5

The file format hasn't changed, so the loading examples below still work. Just replace v12.0 with v12.5 in the filenames. The v12.0 embeddings are still available from the STRING v12.0 archive and Zenodo.

Table of Contents

• Introduction
• Reproduce the Results
• How to Cite
• How to load the embeddings
  • Python Example
  • R Example
  • Read Combined Files
• Contact
• Star History
• License

Introduction

Official repository for the paper in Bioinformatics: SPACE: STRING proteins as complementary embeddings, in which we precalculated:

• cross-species network embeddings
• ProtT5 sequence embeddings

for all eukaryotic proteins in STRING v12.0.

You can download all the embeddings from the STRING website or Zenodo:

• protein.network.embeddings.v12.0.h5
• protein.sequence.embeddings.v12.0.h5

Reproduce the results in the paper

Please follow this document.

How to Cite

If you use this work in your research, please cite the SPACE paper:

Hu, Dewei, et al. "SPACE: STRING proteins as complementary embeddings." Bioinformatics (2025): btaf496. https://doi.org/10.1093/bioinformatics/btaf496

and the STRING database:

Szklarczyk, D., Nastou, K., Koutrouli, M., Kirsch, R., Mehryary, F., Hachilif, R., ... & von Mering, C. (2025). The STRING database in 2025: protein networks with directionality of regulation. Nucleic Acids Research, 53(D1), D730-D737. https://doi.org/10.1093/nar/gkae1113

How to load the embeddings

The following code reads the cross-species network embedding file 9606.protein.network.embeddings.v12.0.h5.

Python example

pip install h5py

import h5py

filename = '9606.protein.network.embeddings.v12.0.h5'

with h5py.File(filename, 'r') as f:
    meta_keys = f['metadata'].attrs.keys()
    for key in meta_keys:
        print(key, f['metadata'].attrs[key])

    embedding = f['embeddings'][:]
    proteins = f['proteins'][:]
	
    # protein names are stored as bytes, convert them to strings
    proteins = [p.decode('utf-8') for p in proteins]

R example:

Install the rhdf5 package to read the embedding files. The following code reads the embedding file 9606.protein.network.embeddings.v12.0.h5.

# Install required packages if not already installed
# install.packages("rhdf5")

# Load the library
library(rhdf5)

filename <- '9606.protein.network.embeddings.v12.0.h5'

metadata <- h5readAttributes(filename, "metadata")
for (key in names(meta_keys)) {
    print(paste(key, meta_keys[[key]]))
}

embeddings <- h5read(filename, "embeddings")
proteins <- h5read(filename, "proteins")

Read combined files

Read the combined network embedding file of all eukaryotes with Python

import h5py

filename = 'protein.network.embeddings.v12.0.h5'

with h5py.File(filename, 'r') as f:
    meta_keys = f['metadata'].attrs.keys()
    for key in meta_keys:
        print(key, f['metadata'].attrs[key])
  
    species = '4932'  # if we check the brewer's yeast
    embeddings = f['species'][species]['embeddings'][:]
    proteins = f['species'][species]['proteins'][:]
	
    # protein names are stored as bytes, convert them to strings
    proteins = [p.decode('utf-8') for p in proteins]

Read the combined file with R

library(rhdf5)

filename <- 'protein.network.embeddings.v12.0.h5'

meta_keys <- h5attributes(h5file$metadata)
for (key in names(meta_keys)) {
    print(paste(key, meta_keys[[key]]))
}

species <- '4932'  # for brewer's yeast
embeddings <- h5read(filename, paste0('species/', species, '/embeddings'))
proteins <- h5read(filename, paste0('species/', species, '/proteins'))

Contact

dewei.hu@sund.ku.dk.

Star history

License

MIT.