Machine learning over interaction networks and ontology-based background knowledge predicts drug--target interactions
Motivation: In silico drug--target interaction (DTI) prediction is important for drug discovery and drug repurposing. Approaches to predict DTIs can proceed indirectly, top-down, using phenotypic effects of drugs to identify potential drug targets, or they can be direct, bottom-up and use molecular information to directly predict binding potentials. Both approaches can be combined with information about interaction networks.
Results: We developed DTI-Voodoo as a computational method that combines molecular features and ontology-encoded phenotypic effects of drugs with protein--protein interaction networks, and uses a graph neural networks to predict DTIs. We demonstrate that drug effect features can exploit information in the interaction network whereas molecular features do not. DTI-Voodoo is designed to predict candidate drugs for a given protein; we use this formulation to show that common DTI datasets contain intrinsic biases with major affects on performance evaluation and comparison of DTI prediction methods. Using a modified evaluation scheme, we demonstrate that DTI-Voodoo improves substantially over state of the art DTI prediction methods.
Python 3.7 packages:
- pytorch 1.6+
- pytorch-geometric 1.6+ (please refer to https://pytorch-geometric.readthedocs.io/en/latest/notes/installation.html)
- numpy 1.19+
- scikit-learn
- networkx
- gensim
- rdflib
- BioPython
- tqdm (for better evaluation and sanity preservation)
Others:
- Groovy (Groovy Version: 2.4.10 JVM: 1.8.0_121) with Grape for dependency management (http://docs.groovy-lang.org/latest/html/documentation/grape.html) for DL2vec axiom generation
- diamond & blastp (for DeepGO feature generation)
Example commands for all used datasets are provided in the src/ folder.
We provide all used datasets and download instructions in the data/ and associated data preparation methods in the src/ directory.
If you find our work useful, please cite:
Tilman Hinnerichs, Robert Hoehndorf, DTI-Voodoo: machine learning over interaction networks and ontology-based background knowledge predicts drug–target interactions, Bioinformatics, 2021;, btab548, https://doi.org/10.1093/bioinformatics/btab548