Name: Jianmin Wang
Type: User
Company: Yonsei University
Bio: Drug Design , Linux enthusiast , Medicinal_Chemistry_&_ Synthesis , Chemoinformatics , Data Science, Python and C/C++ programmer,Bioinformatics,Deep Learning,AI
Twitter: Jianmin4drugai
Location: **(China)
Blog: https://jianmin2drugai.github.io/
Jianmin Wang's Projects
Benchmarks for generative chemistry
"Where files live" - Simple object management system using AWS S3 and Elasticsearch Service to manage objects and their metadata
A series of Jupyter notebooks that walk you through the fundamentals of Machine Learning and Deep Learning in python using Scikit-Learn and TensorFlow.
Cardiovascular diseases incidence probability estimation model
Cardiovascular diseases incidence probability estimation model
RDKit code for the JCIM article
Hierarchical Inter-Message Passing for Learning on Molecular Graphs
Hit-likeness estimation
A Bio2BEL package for converting the Human Metabolite Database (HMDB) to BEL
Aimed to generate new molecules from multiple data set inputs
Heterogeneous Predictors Ensembling for Quantitative Toxicity Prediction
High throughput molecular dynamics simulations
In recent times, toxicological classification of chemical compounds is considered to be a grand challenge for pharma-ceutical and environment regulators. Advancement in machine learning techniques enabled efficient toxicity predic-tion pipelines. Random forests (RF), support vector machines (SVM) and deep neural networks (DNN) are often ap-plied to model the toxic effects of chemical compounds. However, complexity-accuracy tradeoff still needs to be ac-counted in order to improve the efficiency and commercial deployment of these methods. In this study, we implement a hybrid framework consists of a shallow neural network and a decision classifier for toxicity prediction of chemicals that interrupt nuclear receptor (NR) and stress response (SR) signaling pathways. A model based on proposed hybrid framework is trained on Tox21 data using 2D chemical descriptors that are less multifarious in nature and easy to calcu-late. Our method achieved the highest accuracy of 0.847 AUC (area under the curve) using a shallow neural network with only one hidden layer consisted of 10 neurons. Furthermore, our hybrid model enabled us to elucidate the inter-pretation of most important descriptors responsible for NR and SR toxicity.
Benchmark interpretation of QSAR models
Learning Multimodal Graph-to-Graph Translation for Molecular Optimization (ICLR 2019)
Junction Tree Variational Autoencoder for Molecular Graph Generation (ICML 2018)
Implementation of JTVAE using Python 3.6.7, PyTorch 1.0 & Rdkit 2018.09
idock is a standalone tool for structure-based virtual screening powered by fast and flexible ligand docking.
ImageMol is a molecular image-based pre-training deep learning framework for computational drug discovery.
Processing high-throughput drug indication resources.
Universal cheminformatics libraries, utilities and database search tools
Inner- and Outer Recursive Neural Networks for Chemoinformatics
python code for the generation of multi-epitope vaccine constructs
Creating a heterogeneous network for drug repurposing
AAAI 2020 - InteractE: Improving Convolution-based Knowledge Graph Embeddings by Increasing Feature Interactions
Interpreting graph convolutional filters of target prediction network