Cofolding (Beta) Panel

Generate an ensemble of prepared protein-ligand complex structures using multiple cofolding and ML-based structure prediction methods via LiveDesign ML.

To open this panel, do one of the following:

  • Click the Tasks button and browse to Protein Preparation and Refinement → Cofolding (LiveDesign) (Beta)
  • Click the Tasks button and browse to LiveDesign ML Modeling → Cofolding (Beta)

Using the Cofolding (Beta) Panel

Cofolding methods predict protein-ligand complex structures directly from a protein sequence and a ligand structure. Different methods can produce different binding modes, so combining poses from multiple methods produces a more structurally diverse ensemble than relying on any single method. A diverse ensemble increases the chance of finding an accurate binding pose for downstream calculations such as FEP+.

To run a cofolding job:

  1. Connect to LiveDesign using the Connect to LiveDesign Dialog Box.

  2. Provide a FASTA file containing the target protein sequence. FASTA files can be obtained from sequence databases such as UniProt.

  3. Provide a ligand structure. The ligand should be a prepared 3D structure.

  4. Optionally, provide custom template structures to bias structure prediction toward a preferred conformation, or a pocket ASL to reduce search space and filter output poses by the binding site.

  5. Leave Enable fast mode selected for a quicker run with fewer poses, or clear it for a more comprehensive ensemble.

  6. Click Run to submit the job to LiveDesign ML.

The output is an ensemble of aligned protein-ligand complex structures, deduplicated by RMSD and prepared at pH 7.4 using the Protein Preparation Wizard. The structures are imported into the Project Table as a group, with pose interaction analysis, structural diversity analysis, and scoring properties to help you select models for refinement with IFD-MD or validation with FEP+. When fast mode is disabled, output complex structures are grouped by binding site location to facilitate visual screening of poses.

If the connection to LiveDesign is interrupted or you close Maestro while a job is running, Maestro resumes checking the job's status automatically the next time you open the project and reconnect to LiveDesign.

Cofolding (Beta) Panel Features

FASTA file text

Use the Browse button to navigate to and select a FASTA file containing the target protein sequence. The name of the selected file is displayed. FASTA files can be obtained from sequence databases such as UniProt. Supported file formats are .fasta, .fas, .fa, .fst, and .seq.

Ligand file option menu

Choose the source for the ligand structure. Only a single ligand structure is supported. The ligand should be a prepared 3D structure.

  • Project Table (1 selected)—Use the entry that is currently selected in the Project Table. Only one entry should be selected. When this option is selected, a Load button is displayed to the right.

  • File—Use the specified file. When this option is selected, a Browse button is displayed, which you can click to navigate to and select a file.

Custom templates (optional) option menu

Choose the source for custom template structures to bias protein structure prediction toward a preferred conformation. Custom templates are existing protein or protein-ligand complex structures, such as experimental structures from the PDB or prior modeling results. Provide up to 4 structures in Maestro format (.maegz).

  • Project Table (n selected)—Use the entries that are currently selected in the Project Table. The number of entries selected is shown on the menu item. Select up to 4 entries. When this option is selected, a Load button is displayed to the right.

  • File—Use the specified file in Maestro format (.maegz). When this option is selected, a Browse button is displayed, which you can click to navigate to and select a file.

Pocket ASL (optional) text box

Specify pocket residues using chain name and residue numbers in ASL format. When a pocket ASL is provided, cofolding models can use it to reduce search space and output poses are filtered to retain only those with the ligand positioned near the specified binding site. Chains are assigned alphabetically (A, B, C, and so on) based on the order of sequences in the FASTA file, and residue numbering starts at 1 for each chain. For example: ((chain.name A) AND (res.num 1,2)) OR ((chain.name B) AND (res.num 5,6)).

Enable fast mode option

Select this option to generate fewer poses using only cofolding models and return output predictions without post-processing. Selected by default. Clear this option to run the full workflow, which adds homology models from PDB templates and ML-based docking poses for a more structurally diverse ensemble.

Run button

Click Run to submit the job to LiveDesign ML.

Status bar

The status bar displays information about the current job settings and status for the panel. The settings includes the job name, task name and task settings (if any), number of subjobs (if any) and the host name and job incorporation setting. The job status can include messages about job start, job completion and incorporation.

Use the Reset button to reset the panel to its default settings and clear any data from the panel. You can also reset the panel from the Job toolbar.

The status bar also contains the Help button , which opens the help topic for the panel in your browser. If the panel is used by one or more tutorials, hovering over the Help button displays a button, which you can click to display a list of tutorials (or you can right-click the Help button instead). Choosing a tutorial opens the tutorial topic.