Steps - Fold

To fold a protein, we will use the following amino acid sequence as an example, lets use the insulin:

MALWMRLLPLLALLALWGPDPAAAFVNQHLCGSHLVEALYLVCGERGFFYTPKT

Follow these steps to fold the protein:

1

Go to the fold section and select "Protein" as the sequence type.

Usually, "Protein" is selected by default in the form.

2

Copy the insulin protein from the documentation and paste it into the text field.

In the interface, you will also see that insulin is set as the default example.

3

Select the model you will use to run the sequence. (Example: Boltz - 1)

Remember that for proteins, the available models are Boltz - 1, Monomer-ptm-colabfold, Multimer-v3-colabfold, and Solosep-esm-meta, while for DNA, RNA, and ligands, only Boltz - 1 is available.

4

Press the call-to-action button "Fold sequences".

This will open a modal with several options that you can customize for execution.

5

The confirmation modal will appear, where you can optionally add or remove configurations.

By confirming, the sequence will enter an initial "Pending" state, with processing time depending on the number of requested foldings. Below, we show you the state flow of this process.

States flows

When the result is Complete, it means the results are ready to be reviewed by accessing the corresponding job.

If the result is Error, it is recommended to relaunch the job.

In these cases, it is also advisable to contact our support team to ensure a better experience and resolve any issues.

6

Now that you have access to the results

You can view the details of the executed Job or Jobs, along with various graphs that assess the sequence quality. These visualizations will help you analyze the folding performance and validate the accuracy of the obtained results.

To better understand the results, we will first review some essential basic concepts. Then, we will guide you step by step on how to analyze them.

---