/tools

Geo-DEG is a tool designed for data-efficient molecular property prediction by utilizing a learnable hierarchical grammar that generates molecules. It employs graph neural networks to predict properties based on the induced geometry of molecular graphs, outperforming various baseline models.

The TABASCO repository presents a fast and simplified model for generating molecules with improved physical quality. It utilizes a non-equivariant Transformer architecture to treat molecular generation as a sequence modeling problem, making it efficient for generating small molecules.

V2DB is a database and tool for generating and predicting properties of novel two-dimensional materials through virtual screening. It utilizes machine learning models to predict various material properties, facilitating the discovery of stable and functional 2D materials.

Molify is a Python package that facilitates the creation of atomistic structures using RDKit and ASE, allowing users to convert SMILES strings into 3D molecular representations. It also integrates with Packmol to build periodic boxes of molecules, making it useful for molecular simulations.

The smi2sdf3d repository provides a method for generating diverse 3D conformers of molecules using the RDKit library. This tool is useful for researchers in computational chemistry and drug discovery, facilitating the exploration of molecular conformations.

Pseudocycle_small_molecule_binder is a tool for the de novo design of diverse small molecule binders using shape complementary pseudocycles. It provides scripts and notebooks for generating and optimizing molecular structures, specifically targeting small molecules for binding applications.

`augment-atoms` is a tool designed for augmenting datasets of atomic configurations using a model-driven approach. It generates new molecular structures by applying transformations to existing ones, making it useful for enhancing datasets in molecular machine learning applications.

ProCALM is a protein conditionally adapted language model designed for the generation of enzymes based on specific conditions such as EC number and taxonomy. It utilizes autoregressive transformers to create functional sequences, making it a valuable tool for protein design and molecular generation.

READRetro is a tool designed for natural product biosynthesis planning using retrieval-augmented dual-view retrosynthesis. It allows users to evaluate and plan retrosynthesis paths for various molecules, making it relevant for molecular design and generation tasks.

GenUI is a frontend application that offers a graphical user interface for interacting with the GenUI REST API web services. It is designed to facilitate molecular generation and cheminformatics tasks through a user-friendly dashboard.

AMix-1 is a protein foundation model that employs a test-time scalable approach to generate and evaluate protein sequences. It allows for the iterative evolution of protein designs through a systematic evaluation of various metrics, making it a valuable tool for protein engineering.

DrugGen is a tool that enhances drug discovery by using large language models to generate drug-like SMILES structures from protein sequences. It employs reinforcement learning and supervised fine-tuning to ensure the generated structures are chemically valid and effective.

Drug_Design_Models is a reimplementation of various models for de novo drug design, utilizing techniques such as recurrent neural networks and graph convolutional networks. It aims to generate and optimize molecular structures based on established research in the field.

The mlx-esm repository provides an implementation of Meta's ESM-1 protein language model using the MLX library. It allows users to generate new protein sequences and predict masked amino acids, contributing to protein design and structure prediction in molecular biology.

PMGen is a comprehensive pipeline designed for predicting peptide-MHC complex structures and optimizing peptide sequences. It utilizes advanced techniques like AlphaFold for structure prediction and includes features for iterative peptide optimization and mutation screening.

iPPIGAN is a tool for de novo molecular design that utilizes deep molecular generative models to create inhibitors targeting protein-protein interactions. It includes functionalities for training models and generating novel compounds, contributing to drug discovery efforts.

PoET-2 is a multimodal protein language model designed for generating protein sequences and predicting the effects of protein variants. It utilizes retrieval-augmented techniques to enhance its predictive capabilities, making it a valuable tool in protein design and bioinformatics.

This repository provides tutorials and pipelines for the computational design of metallohydrolases using the RFdiffusion2 model. It includes both dry lab data and wet lab data, facilitating the design and testing of enzyme sequences.

DrugPilot is an advanced LLM-based agent framework that automates and optimizes various aspects of drug discovery. It includes functionalities for predicting drug properties, generating new drug candidates, and classifying drug-target affinities, thereby streamlining the drug discovery process.

The DrugHunting repository provides Python scripts for automating drug discovery processes, including the design and optimization of drug-like molecules. It utilizes stochastic methods and cheminformatics to explore novel chemical spaces, making it suitable for applications like docking and virtual screening.

CCMgen and CCMpredPy provide a Python toolkit for generating realistic synthetic protein sequences using second-order Markov Random Field models. The tools allow for the sampling of protein-like sequences while adhering to evolutionary pressures, making them useful for protein design and analysis.

The paccmann_sarscov2 repository provides a pipeline for automating the discovery and synthesis of targeted molecules using machine learning. It includes models for predicting protein-ligand interactions, toxicity, and generative models for both proteins and small molecules.

Syndirella is a tool designed for generating and scoring synthetically tractable elaborations of molecules derived from fragment screens. It utilizes retrosynthetic analysis and energy minimization to produce viable synthetic routes for small molecules.

JAMUN is a tool that bridges smoothed molecular dynamics and score-based learning to efficiently generate conformational ensembles of peptides. It enhances the sampling of molecular dynamics by operating in a smoothed space, allowing for faster and more transferable results in molecular simulations.