/tools

MolDesigner is an interactive web interface that helps drug developers design efficacious drugs using deep learning predictions. It allows users to input drug molecules and receive real-time predictions on important indices related to drug efficacy, facilitating iterative design.

This repository implements algorithms for link prediction in a Chemical Reaction Knowledge Graph, enabling the prediction of novel reactions and products. It utilizes models like SEAL and Graph Auto-Encoder to facilitate the design and generation of chemical compounds.

AdaptivePELE is a Python package that enhances the sampling of molecular simulations using the Protein Energy Landscape Exploration method. It allows users to perform adaptive sampling, clustering, and trajectory analysis for better modeling of protein-ligand interactions.

Etomica is a molecular simulation framework developed in Java, designed for simulating molecular systems. It provides tools for modeling and analyzing molecular dynamics, making it relevant for researchers in computational chemistry and molecular biology.

MoSDeF Workflows is a collection of sample workflows that utilize the Molecular Simulation Design Framework to support reproducible molecular simulations across various engines. It includes tools for building molecules, applying forcefields, and managing simulation data, making it suitable for molecular dynamics and simulations.

GraSeq is a tool designed for predicting molecular properties through graph and sequence fusion learning. It provides implementations for both single-task and multi-task classification of various molecular datasets, making it a valuable resource for researchers in computational chemistry.

NHGNN-DTA is a tool designed for predicting drug-target binding affinity using a node-adaptive hybrid graph neural network. It includes functionalities for data preparation, model training, and evaluation, making it relevant for molecular property prediction in the context of drug discovery.

DrugAI_Drug-Likeness is a tool that evaluates the drug-likeness of molecules based on various molecular properties and structure features. It implements several rules and filters, such as Lipinski's Rule of Five, to assess the suitability of compounds for drug development.

AtomsCalculators.jl is a Julia interface designed for atomistic calculations, focusing on molecular dynamics and geometry optimization. It aims to extend functionalities for calculating energies and forces, with potential for broader applications in molecular simulations.

TransAntivirus is a transformer-based molecular generative model aimed at designing antiviral drugs. It allows for the generation of novel compounds and is built upon existing frameworks for molecular generation, making it a valuable tool in the field of drug discovery.

NAG2G is a neural network model designed for predicting retrosynthesis pathways in molecular chemistry. It supports enhanced stereochemistry features and provides datasets and pretrained weights for effective model validation and usage.

The openprotein-python repository offers a user-friendly interface for the OpenProtein.AI API, enabling users to perform tasks related to protein analysis, including sequence generation and scoring using generative models. It supports various functionalities for protein modeling and design, making it a valuable tool in the field of molecular biology.

S-CGIB is a pre-training architecture for Graph Neural Networks aimed at predicting molecular properties without human annotations. It utilizes self-supervised learning to generate graph-level representations and has been tested on various molecular datasets.

LABind is a structure-based method that predicts binding sites of proteins with ligands, focusing on interactions between ligands and proteins. It utilizes machine learning techniques to enhance the accuracy of binding site predictions.

AttentionDTA_TCBB is a tool designed for predicting drug-target binding affinities using a sequence-based deep learning model with an attention mechanism. It includes datasets for training and testing the model, making it relevant for molecular property prediction in drug discovery.

SynthArena is an interactive platform designed for visualizing and comparing retrosynthetic routes generated by AI models. It facilitates the evaluation of these models through standardized benchmarks and provides features for side-by-side comparison of synthetic routes.

SCEPTR is a transformer-based model designed for T cell receptor (TCR) representation, enabling alignment-free analysis and prediction of TCR-pMHC interactions. It outperforms traditional methods in TCR specificity prediction, making it a valuable tool in immunoinformatics.

CATHe is a deep learning tool that utilizes protein sequence embeddings to detect remote homologues for CATH superfamilies. It achieves high accuracy in predicting protein classifications, making it a valuable resource for researchers in molecular biology and bioinformatics.

RNAvigate is a toolset designed for exploring and comparing chemical probing data and structure models of RNA. It offers an easy-to-use interface, particularly effective in Jupyter Notebooks, for documenting and sharing analyses related to RNA structures.

The Therapeutic Nanobody Profiler (TNP) is an open-source computational tool that characterizes and predicts the developability of nanobodies to enhance therapeutic design. It utilizes unique metrics tailored for nanobodies, based on experimental data and clinical-stage sequences, to facilitate their development as biotherapeutics.

GlyLES is a Python tool that converts IUPAC representations of glycans into SMILES strings, facilitating the representation and analysis of glycan structures in computational chemistry. It supports various input formats and is currently in development to enhance its functionality.

QuickSES is a tool designed to compute molecular Solvent Excluded Surface meshes using GPU acceleration with CUDA. It allows users to generate high-quality surface representations of molecules from PDB files, facilitating molecular visualization and analysis.

DeepAllo is a deep learning framework designed for predicting allosteric sites in proteins using a protein language model with multitask learning. It provides an inference pipeline that identifies potential allosteric pockets based on input protein structures.