/tools

ConfVAE-ICML21 is an end-to-end framework designed for generating molecular conformations using a variational autoencoder approach. It provides tools for training models on molecular datasets and generating conformations for various molecules, making it a valuable resource in computational chemistry and molecular design.

SGGRL is a framework designed for multi-modal representation learning aimed at predicting molecular properties using sequence, graph, and geometry data. It provides functionalities for processing data and performing molecular property predictions, making it relevant for computational chemistry applications.

CARBonAra is a deep learning framework that facilitates the design of protein sequences by utilizing atomic coordinates and context-aware generation methods. It allows for the integration of molecular environments, including non-protein entities, enhancing the control over protein engineering and design.

The S3F repository provides a multimodal representation learning framework for predicting protein fitness. It integrates various protein features and is evaluated using the ProteinGym benchmark, making it a valuable tool for protein design and property prediction.

Platonic Transformers is a framework that integrates geometric group theory into transformer architectures to enhance molecular property prediction. It supports various molecular datasets, including QM9 for quantum chemistry properties and OMol for molecular learning tasks.

AutoTST is a framework that automates transition state theory calculations for reaction families commonly found in combustion chemistry. It utilizes existing quantum chemistry packages to optimize molecular geometries and obtain kinetic parameters.

Jrystal is a JAX-based framework designed for differentiable density functional theory calculations, enabling efficient optimization workflows for quantum properties of materials. It supports solid-state calculations and is optimized for GPU performance, making it suitable for advanced electronic structure computations.

MapDiff is a PyTorch implementation of a deep diffusion model designed to improve the inverse protein folding task. It predicts feasible amino acid sequences from 3D protein backbone structures, contributing to the field of protein design.

dZiner is an AI framework designed for the rational inverse design of materials, allowing for the generation and assessment of new molecular candidates based on user-defined properties and constraints. It incorporates a human-in-the-loop approach to refine designs and improve chemical feasibility.

I-ReaxFF is a differentiable framework for the Reactive Force Field (ReaxFF) based on TensorFlow, allowing for the optimization of ReaxFF parameters using machine learning techniques. It facilitates molecular simulations by providing first and higher-order derivatives of energies, which are essential for various computational chemistry applications.

The KIM API is a system-level library designed to facilitate the development of atomistic and molecular simulation programs by providing a standardized interface for various interatomic models. It supports multiple programming languages, enabling seamless integration into existing simulation workflows.

Uni-MOF is a transformer-based framework designed for high-accuracy predictions of gas adsorption in metal-organic frameworks (MOFs). It utilizes a large dataset of MOF structures to learn representations and predict various properties, making it a valuable tool in computational chemistry and materials science.

BLIP is a framework for training and fine-tuning machine learning interatomic potentials, providing reliable uncertainty estimates and minimal computational overhead. It allows users to convert existing machine learning interatomic potentials into Bayesian models, enhancing their robustness and interpretability.

DIMOS (Differentiable Molecular Simulator) is a PyTorch-based framework that enhances molecular dynamics and Monte Carlo simulations through machine learning. It allows for the integration of classical force fields and machine learning interatomic potentials, facilitating advanced research in computational chemistry and biology.

RAG-ESM is a framework that enhances pretrained protein language models by conditioning them on homologous sequences. It allows for the generation of novel protein sequences and improves predictive performance through a retrieval-augmented approach.

IDSL_MINT is a deep learning framework designed to interpret raw mass spectrometry data, enabling the prediction of molecular fingerprints and structures from MS/MS spectra. It utilizes transformer models to translate mass spectra into molecular descriptors and canonical SMILES, making it a valuable tool for cheminformatics and molecular property prediction.

LightDock is a docking framework that utilizes the Glowworm Swarm Optimization algorithm to facilitate protein-protein, protein-peptide, and protein-DNA docking. It allows users to define custom scoring functions and supports various simulation options, making it a versatile tool for molecular docking studies.

The Automated QSAR Framework is designed for the curation of chemogenomics data and the development of predictive QSAR models using machine learning. It facilitates data preparation, chemical space analysis, and virtual screening, making it a valuable tool for drug discovery.

HimGNN is a PyTorch implementation of a novel hierarchical molecular representation learning framework aimed at predicting molecular properties. It utilizes a combination of Atom-MPNN and Motif-MPNN to enhance the performance of molecular property prediction tasks through advanced graph-based techniques.

OPUS-Fold is an open-source framework designed for protein folding based on torsion-angle sampling. It integrates various methods for protein structure prediction and allows for efficient modeling of protein side chains during the folding process.

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.

BC-Design is a framework designed for high-precision inverse protein folding, integrating structural and biochemical features to enhance protein design accuracy. It utilizes a dual-encoder architecture to generate amino acid sequences that correspond to specific 3D protein structures, making it valuable for protein engineering and drug development.

The MDAnalysis Toolkits Registry is a repository that hosts and documents various MDAKits, which are tools designed for molecular simulation analysis. It aims to provide a structured approach to developing and validating tools relevant to molecular dynamics and computational chemistry.

Reciprocal Space Attention (RSA) is a machine learning framework designed to capture long-range interactions in molecular systems by utilizing a Fourier domain approach. It enhances existing interatomic potentials by addressing the limitations of local and semi-local models, making it suitable for various chemical and materials systems.