Web Analytics Made Easy - Statcounter
Skip to content

Predict Biomolecular Structures#

Thanks to the Biomolecular Structure Prediction extension in SAMSON, you can predict biomolecular structures using one of the following services:

  • AlphaFold-2
  • Boltz-2
  • Chai-1

You can access the prediction services through Home > Predict.

The predictions are done in the cloud, and data are sent via a secure connection. You can view the results in SAMSON in Interface > Cloud jobs or on SAMSON Connect > Account > Jobs.

Note

When prediction results are loaded in SAMSON, it will colorize the structures based on pLDDT values, if they are available in the files.

Predicting with AlphaFold-2#

Predict biomolecular structures using AlphaFold-2 1 in a few steps in SAMSON:

  • Open Home > Predict.
  • Choose AlphaFold-2 service.
  • Enter one or more FASTA files.
  • Choose the AlphaFold model (monomer, multimer, etc.) and the database for multiple sequence alignment.
  • Click Start prediction.

Note

Predictions are performed using a choice of cloud machines with varying performance and costs, including powerful instances with A100 GPUs.

Contact us to request computing credits or buy computing credits directly.

Any publication that discloses findings arising from using the AlphaFold service should cite the AlphaFold paper 1 and, if applicable, the AlphaFold-Multimer paper.

Predicting with Boltz-2#

Predict biomolecular structures using Boltz-2 2 in a few steps in SAMSON:

  • Open Home > Predict.
  • Choose Boltz-2 service.
  • Click on Add protein, Add DNA, Add RNA, or Add ligand to enter the corresponding sequences, CCD codes, or ligand SMILES strings.
  • Click Add modification if you want to alter the sequences (for proteins, DNA, or RNA), and enter the position (the residue index) and the CCD code of the modification.
  • Click Start prediction.

Note

Calculations are performed on a dedicated A100 GPU in the cloud and use computing credits. A typical structure prediction with Boltz-2 will cost between 0.5 and 1 computing credit.

Contact us to request computing credits or buy computing credits directly.

You can watch a video tutorial and learn how to:

Predicting with Chai-1#

Predict biomolecular structures using Chai-1 3 in a few steps in SAMSON:

  • Open Home > Predict.
  • Choose Chai-1 service.
  • Click on Add protein, Add DNA, Add RNA, or Add ligand to enter the corresponding sequences or SMILES strings.
  • Click Add pocket restraint if you want to add a distance restraint between a residue (for proteins, DNA, or RNA) and a chain.
  • Click Add contact restraint if you want to add a distance restraint between two residues.
  • Click Start prediction.

Note

Calculations are performed on a dedicated A100 GPU in the cloud and use computing credits. A typical structure prediction with Chai-1 will cost between 0.5 and 1 computing credit.

Contact us to request computing credits or buy computing credits directly.

Need Help?#

Have questions or feedback? Feel free to reach out via the Forum, via e-mail, via the Feedback button in SAMSON, or by directly discussing with us.

References#

  1. Jumper, J., Evans, R., Pritzel, A. et al. Highly accurate protein structure prediction with AlphaFold. Nature 596, 583–589 (2021). https://doi.org/10.1038/s41586-021-03819-2 ↩↩

  2. Saro Passaro, et al. Boltz-2: Towards Accurate and Efficient Binding Affinity Prediction. bioRxiv 2025.06.14.659707 ↩

  3. https://www.chaidiscovery.com/blog/introducing-chai-1 ↩