Check out the latest news for Nanome AI Early Access. Learn More
Back to Blog

Computer-Aided Drug Discovery & Design through MARA

August 21, 2024

Computer-Aided Drug Discovery & Design through MARA

Reduce the steps you take to complete your daily tasks

With the release of MARA Public Beta, we are thrilled to showcase the incredible enterprise ready capabilities of this powerful tool. Workflows for Computer Aided Drug Discovery & Design (CADDD) and Structural Based Drug Design (SBDD) projects require an investigation of three-dimensional structures of biological macromolecules, such as proteins, nucleic acids, and carbohydrates to understand their function and interactions at the molecular level. Scientists completing these tasks are critical to molecular modeling, data analysis, experimental design, and structural determination. All of this work is essential for drug discovery, understanding disease mechanisms, developing new therapeutic strategies.

Here, we outline a simple enterprise ready MARA workflow for a CADDD/SBDD scientist, illustrating how MARA can save you valuable time and lighten your workload.

Workflow Overview

  • Load AlphaFold structure
  • Prepare protein structure for docking (including a PDB file fixer tool)
  • Define potential pockets
  • Draw a molecule in 2D and convert to 3D as an SDF file
  • Minimize SDF file and add hydrogens
  • Dock the minimized molecule at the prepared AlphaFold structure at a desired pocket
  • Send to Nanome app Quick Drop to interact with in XR

Although ESMfold can be queried directly from MARA, and a tool can be made to query AlphaFold directly, users may already have AlphaFold CIF files they want to use. An entire workflow to go from a predicted protein structure to a finished docking experiment can be completed using just a few simple natural language prompts. The final product can be downloaded or conveniently sent to a Nanome app Quick Drop for further analysis.

Load an AlphaFold CIF Structure and Prepare for Docking

A prompt of “Prepare for docking” was used when loading the CIF file. MARA then handled the file conversions necessary to produce a PDBQT file and also ran the tools necessary to define pockets within the protein scaffold. For illustrative effects anthracene was drawn in MARA using the molecular drawing tool and imported to MARA as a SMILES string. MARA combines the use of multiple tools from a single prompt “minimize and add hydrogens” to produce an SDF file for docking experiments.

Using Mara to Prepare for DockingUsing Mara to Prepare for Docking

Define Pockets for Docking from a Tabular Output Created During Docking Preparation

During the docking preparation MARA outputs a structure with docking locations and a matching table with pocket id’s and x,y,z coordinates. A prompt of “use the sdf file to dock at the pdbqt file in pocket_id 1” initiated MARA to complete a series of tools to dock the hydrogenated and minimized anthracene molecule at the prepared predicted protein structure.

Using Mara to Dock for Based on PocketsUsing Mara to Dock for Based on Pockets

Send Docking Products to Local Drive or Nanome App Quick Drop

MARA integrates seamlessly into your existing data infrastructure and the Nanome data infrastructure; easily send your MARA produced structures to Nanome to interact with in XR. All files produced by MARA for the user are available to download directly to any device as well.

Using Mara to Send Docked Results to Nanome VaultUsing Mara to Send Docked Results to Nanome Vault

Summary

MARA is here to help guide your enterprise scientific workflows. MARA is adaptive and can iterate through logical steps to find the best possible path. What tools will you create? How much time can you save?

Want to try MARA out yourself? Click here to begin with 50 free credits.