FroGS embeddings generator

The app generates embeddings using the methods developed in:

Chen, H., King, F.J., Zhou, B. et al. Drug target prediction through deep learning functional representation of gene signatures. Nature Commun. 15, 1853 (2024). https://doi.org/10.1038/s41467-024-46089-y

Instructions

1. Upload the GO embedding file (CSV format).
2. Upload the ARCHS4 embedding file (CSV format).
3. Upload your gene signatures text file (TXT format).
4. Click the “Process Files” button to generate embeddings.
5. Once processing is complete, a download button will appear.
6. Click the “Download Embeddings CSV” button to download the generated embeddings file.

Important Notes

• The gene signatures file should be a text file where each line represents a gene signature.
• Each line should start with the signature name, followed by a comma, and then a comma-separated list of genes.
• Ensure that the genes in your list match the gene identifiers used in the embedding files (i.e., NCBI gene IDs).
• The embedding files should be in CSV format with gene IDs in the first column and embedding values in the subsequent columns.

Precomputed GO embeddings

Precomputed ARCHS4 embeddings