/tools

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.

FMol is a simplified drug discovery pipeline that generates SMILE molecular representations using AlphaSMILES, predicts protein structures with AlphaFold, and evaluates druggability using fpocket and Amber. It integrates various molecular modeling techniques to facilitate drug discovery processes.

OpenMM-CpH is a package for performing constant pH simulations using OpenMM, allowing for the exploration of molecular dynamics with varying protonation states. It automates the determination of reference energies for protonation states, facilitating accurate simulations in explicit solvent.

The ESPF tool provides a method for generating explainable substructure partition fingerprints for proteins and drugs, facilitating molecular property prediction. It includes datasets and examples for customizing the fingerprint generation process, making it useful for drug discovery and protein design.

MD-AlanineDipeptide is a tool for performing molecular dynamics simulations of the alanine dipeptide, analyzing its conformational states and energy landscape. It utilizes GROMACS for simulations and provides insights into dihedral angle distributions and transition states.

Longbow automates the submission and monitoring of molecular simulations on remote high-performance computing machines. It supports various bio-molecular simulation software and job schedulers, making it easier for researchers to run complex simulations efficiently.

FS-GNNTR is a few-shot learning framework that utilizes graph embeddings and transformers to predict molecular properties. It addresses low-data challenges in molecular property prediction and demonstrates strong performance on datasets like Tox21 and SIDER.

MoleculeMO is a tool for multiobjective de novo drug design that utilizes recurrent neural networks to generate and optimize molecules based on various properties. It includes data preprocessing, model training, and validation of generated molecules for their pharmacokinetic properties.

The esmjax repository provides a JAX/Flax reimplementation of the ESM-2 protein language model, enabling efficient inference and embeddings for protein sequences. It includes features for model parallelism and weight porting from PyTorch, making it a useful tool for researchers in protein-related computational tasks.

xTrimoPGLM is an open-source family of protein language models developed for tasks related to protein understanding and design. It includes both masked and causal language models, enabling applications in protein sequence analysis and generative design.

PTMGPT2 is a GPT-based model designed to predict post-translational modification sites in protein sequences. It utilizes an autoregressive transformer architecture to generate classification labels for various PTM types, making it a valuable tool for protein analysis.

CiderPress is a high-performance software package designed for training and evaluating machine-learned exchange-correlation functionals within the CIDER framework. It provides interfaces to popular quantum chemistry codes, facilitating advanced computational studies in molecular systems.

GENiPPI is an interface-aware molecular generative framework aimed at designing modulators for protein-protein interactions. It utilizes a dataset of PPI interfaces to generate novel compounds, enhancing the capabilities of structure-based drug design.

DiffIUPAC is a diffusion-based generative model that facilitates molecular editing by converting chemical natural language (IUPAC names) to SMILES strings. It aims to enhance molecular design and optimization, demonstrating applicability in drug design through various case studies.

Uni-Dock-Benchmarks is a repository that contains a curated collection of datasets and benchmarking tests for assessing the performance and accuracy of the Uni-Dock docking system. It includes prepared structures and input files for both molecular docking and virtual screening, making it a valuable resource for researchers in computational chemistry.

The 'nma' repository provides tools for performing Normal Mode Analysis on macromolecules, enabling users to analyze molecular motions and generate trajectories based on specific modes. It is particularly useful for studying the dynamics of proteins and other macromolecular structures.

HyFactor is an open-source architecture designed for generating molecular structures using a novel Hydrogen-count Labelled Graph approach. It includes implementations for both HyFactor and ReFactor architectures, facilitating the translation between molecular graphs and their representations.

CurrMG is a tool that implements a curriculum learning-based strategy for molecular graph learning, enabling the prediction of molecular properties using various graph neural network models. It supports multiple datasets and evaluation metrics, making it suitable for molecular property prediction tasks.

UQ4DD is a Python package designed for uncertainty quantification in drug discovery, specifically for predicting molecular properties and drug-target interactions. It utilizes deep learning techniques to provide estimates of uncertainty alongside property predictions, making it a valuable tool for researchers in the field.

This repository provides an introductory workshop on protein language models, exploring the theoretical and practical aspects of protein embeddings and their applications in protein-related machine learning tasks.

The Phylogeny-MSA-Transformer repository provides tools for training protein language models on multiple sequence alignments to learn phylogenetic relationships. It includes scripts and requirements for analyzing protein sequences and generating phylogenetic insights.

The 'openmm_scripts' repository provides a collection of Python scripts that facilitate the setup and execution of molecular dynamics simulations with OpenMM. It is aimed at users who need to perform simulations in the field of biophysics and molecular dynamics.

Pose is a Python library that allows users to construct and manipulate protein molecular structures, including building polypeptides from sequences and performing various structural manipulations. It provides functionalities for analyzing molecular properties such as potential energy and radius of gyration, making it a valuable tool for protein design and molecular simulations.