Features

Chat

This is the primary interface for interacting with MARA. It is a chat interface that can help you analyze and reason about molecular data by constructing workflows of tools to achieve a specific goal.

Here are some examples of the types of tasks MARA can accomplish:

• Data Retrieval:
  • Downloading PDB files for specific proteins.
  • Retrieving SMILES strings for given compounds.
  • Extracting protein sequences from PDB or mmCIF files.
• Data Conversion:
  • Converting PDB files to mmCIF files and vice versa.
  • Converting SMILES strings to 3D structures.
  • Converting molecular dynamics (MD) trajectories to PDB files.
• Informatics:
  • Calculating chemical descriptors for lists of SMILES.
  • Finding similar compounds in databases like ChEMBL using SMILES strings.
  • Generating molecular fingerprints from SMILES.
• Analysis:
  • Aligning reference polymer structures to sets of mobile structures.
  • Computing solvent accessible surface area (SASA) of proteins.
  • Calculating RMSD from MD trajectories.
• Visualization:
  • Drawing 2D structures of molecules from SMILES strings.
  • Generating 3D conformers for given molecules.
  • Displaying protein sequences and alignments visually.

Chat Input

Along with natural language, prompts can be augmented with files, or a SMILES can be drawn using a Ketcher chemical structure editor.

• Hexagon button opens the chemical structure drawing tool

• Upload buttons opens a file picker to select files to upload

Data Table

With the MARA data table, you can perform a variety of tasks to analyze and manipulate your data.

Here are some of the key functionalities:

• Query and Extract Data from Table: You can ask natural language queries to extract specific data from a CSV file. This is useful for quick insights without manual analysis.
• Generate a 2D or 3D Scatter Plot: You can create scatter plots from your CSV data, which can be either 2D or 3D. This helps in visualizing the relationships between different data points.
• Export Data to CSV File: You can write data into a CSV file, creating or overwriting it as needed. This is ideal for exporting processed or analyzed data into a CSV format.
• Filter and Extract CSV Data Based on a Query: You can filter and extract specific rows from a CSV file based on a query, allowing you to focus on the most relevant data.
• Show Distributions of Numerical Data: You can visualize the distribution of numerical data from a CSV file, which can help in understanding the spread and central tendencies of your data.
• Understand CSV File Content: You can get a summary or overview of the contents of a CSV file, which can be useful for initial data exploration.

These tools provide a comprehensive suite for handling and analyzing data tables, making it easier to derive meaningful insights from your datasets.

Creating Tables

You can create a table by uploading a CSV file through the Chat input, or by dragging and dropping a file into the interface.

The drag and drop interface contains two drop zones; the left side will upload the file to the chat, and the right side will upload the file and create a new table using that data.

Supported file types for tables: .csv, .xlsx

Molecular Visualization

MARA allows you to visualize structures in 3D using Molstar, and update the representation using natural language.

Example Prompts:

• Show all residues in chain D as ball and stick
• Color residues 101-207 blue
• Hide waters
• Spin the molecule
• Focus on ligand 50D
• Download as [ molx| stl | obj | glb | usdz ]
• Highlight residues 149-150 in green
• Set style to illustrative