Colabfold Batch

This node is currently disabled.

Need access? Contact us and we’ll help you out.

Runs Colabfold model inference on Colabfold MSA search results, generating 5 ranked PDB foldings per input sample.

Quick Start

1. Prepare MSA search results in A3M format using the Colabfold Search node.
2. Connect the A3M output to the Colabfold Batch node.
3. Run the node to generate ranked PDB foldings.

Setup Guide

1. Prepare Input Data

1. Use the Colabfold Search node to obtain MSA results in A3M format.
2. Ensure each sequence is properly formatted and identified.

2. Run Colabfold Batch

1. Connect the A3M input to the Colabfold Batch node.
2. Execute the node to receive ranked PDB foldings for each input sequence.

Basic Usage

Batch Structure Prediction

• Input a batch of A3M-formatted MSA results.
• Obtain 5 ranked PDB foldings per input sample.
• Use outputs for downstream analysis or visualization.

Configuration

Required Inputs

Field	Description	Type	Example
a3m	MSA search results in a3m format.	A3M	{"seq1": "...", "seq2": "..."}

Optional Inputs

None

Outputs

Field	Description	Type	Example
folding.pdb	Ranked foldings as a dict {pdb_id_rank_id: pdb_content}.	PDB	{"seq1_ranked_1": "...", "seq1_ranked_2": "..."}

Best Practices

Input Preparation

• Ensure A3M input is generated by the Colabfold Search node for compatibility.
• Use clear, unique sequence IDs to avoid confusion in output mapping.

Output Handling

• Use the ranked outputs for further evaluation or visualization.
• Integrate with downstream nodes for workflow automation.

Troubleshooting

Common Issues

• No output generated: Verify that the A3M input is correctly formatted and not empty.
• Unexpected output IDs: Ensure input sequence IDs are unique and consistent.

Need Help?

• Contact support for further assistance.