DeepDrug Protein Embeddings Bank (DPEB)

bioinformatics life sciences machine learning protein structural biology

Description

DPEB is a multimodal database of human protein embeddings integrating four biologically complementary representations—AlphaFold2, BioEmbeddings, ESM-2, and ProtVec—designed for enhanced protein-protein interaction prediction and functional classification.

Update Frequency

Initial release; maintained for at least 2 years with updates planned based on new embedding models and protein coverage.

License

MIT

Documentation

https://github.com/deepdrugai/DPEB

Managed By

Louisiana State University

See all datasets managed by Louisiana State University.

Contact

https://github.com/deepdrugai/DPEB/issues

How to Cite

DeepDrug Protein Embeddings Bank (DPEB) was accessed on DATE from https://registry.opendata.aws/deepdrug-dpeb. Sajol MSI et al. DeepDrug Protein Embeddings Bank (DPEB) was accessed on [DATE] at https://registry.opendata.aws/dpeb

Usage Examples

Tutorials
Tools & Applications
Publications

Resources on AWS

  • Description
    Multimodal human protein embeddings (AlphaFold2, BioEmbeddings, ESM-2, ProtVec) with JSONL-formatted metadata containing FASTA, UniProt IDs, and embeddings.
    Resource type
    S3 Bucket
    Amazon Resource Name (ARN)
    arn:aws:s3:::deepdrug-dpeb
    AWS Region
    us-west-2
    AWS CLI Access (No AWS account required)
    aws s3 ls --no-sign-request s3://deepdrug-dpeb/

Edit this dataset entry on GitHub

Tell us about your project

Home