Colab Search API

The Colab Search api runs the MSA generation tool developed by the team at ColabFold on a private server. MSA generation is required step for many ai protein structure prediction tools such as AlphaFold2, OpenFold, RosettaFold, and Boltz. In many cases these MSA files can be precomputed using this tool and then used as input to those other tools. More information can be found on the ColabFold github page

Command Line Interface

Examples

Generate the MSA for input.fasta:

lev engine submit colab-search input.fasta

Flags

  • --fasta-file (str) (Required)
    • Input FASTA file with query sequence
    • You can specify this flag as the first positional argument or with the --pdb-file flag
    • Examples:
      • "input.fasta" - Local FASTA file
      • "sequences/protein.fasta" - FASTA file in subdirectory
      • "/path/to/sequence.fasta" - Full path to FASTA file

Python Interface

Examples

Generate the MSA for input.fasta:

from engine import EngineClient

client = EngineClient()
client.authorize()

job_id = client.submit_colab_search(
    fasta_file="input.fasta"
)

Flags

  • fasta_file (str) (Required)
    • Input FASTA file with query sequence
    • Examples:
      • "input.fasta" - Local FASTA file
      • "sequences/protein.fasta" - FASTA file in subdirectory
      • "/path/to/sequence.fasta" - Full path to FASTA file

Outputs

This API generates a single output file labeled 0.a3m. Which is your MSA in a3m format.

Updated: