Dataset for "Large Language Models as molecular design engines"

1. claude-gpt-paper.zip :
   
   This dataset contains data and results associated with the paper "Large Language Models as molecular design
   engines" The paper investigates the use of large language models, specifically Claude 3 Opus, for generating and analyzing chemical structures based on various prompts from A-H (as mentioned in the manuscript), and guided design related to electron-withdrawing groups (EWG), electron-donating groups (EDG).

The dataset includes:

1. PM7 MOPAC energy calculations for generated molecules, along with their SMILES representations and molecule IDs.
2. PM7-calculated charges for the generated molecules.
3. Output files from the Claude 3 Opus language model for each prompt category along.
4. Original dataset (subset of ZINC database) used to build common keys and the initial design space.
5. JSON file containing common keys for featurizing unknown SMILES.
6. PCA object to convert molecule embeddings to 3-dimensional embeddings.

The data is organized into the following folders:

• pm7_charge_results: Contains HOMO-LUMO energy differences for plotting.
• pm7_charge_calculation: Contains PM7 MOPAC energy calculations and charges.
• out: Contains output files from the Claude 3 Opus language model.
• fact-dropbox: Contains the original dataset, common keys, and PCA object file.

The data can be used to reproduce the results presented in the paper and serve as a foundation for further research in this area.

For a detailed description of the folder structure and contents, please refer to the File_descriptions.md file included in the dataset.


2. llm-visulizer-dashapp.zip

This is the code for the visualizer app for viewing the molecules generated by the LLM. The README.md file has details about running the app.

3. claude-gpt-paper-codes.zip 

This contains the notebook GPT_modification_just_plots.ipynb for plotting, and other codes. The README.md file has details about running the main notebook for getting the plots.