Summary of Mars: a Neurosymbolic Approach For Interpretable Drug Discovery, by Lauren Nicole Delong et al.
MARS: A neurosymbolic approach for interpretable drug discovery
by Lauren Nicole DeLong, Yojana Gadiya, Paola Galdi, Jacques D. Fleuriot, Daniel Domingo-Fernández
First submitted to arxiv on: 2 Oct 2024
Categories
- Main: Artificial Intelligence (cs.AI)
- Secondary: Machine Learning (cs.LG); Logic in Computer Science (cs.LO)
GrooveSquid.com Paper Summaries
GrooveSquid.com’s goal is to make artificial intelligence research accessible by summarizing AI papers in simpler terms. Each summary below covers the same AI paper, written at different levels of difficulty. The medium difficulty and low difficulty versions are original summaries written by GrooveSquid.com, while the high difficulty version is the paper’s original abstract. Feel free to learn from the version that suits you best!
| Summary difficulty | Written by | Summary |
|---|---|---|
| High | Paper authors | High Difficulty Summary Read the original abstract here |
| Medium | GrooveSquid.com (original content) | Medium Difficulty Summary This paper proposes a novel approach called Neurosymbolic (NeSy) artificial intelligence that combines logic or rule-based techniques with neural networks for biomedical applications like drug discovery. NeSy methods are promising due to their enhanced interpretability, which is crucial for understanding the biological plausibility of model interpretations. The authors develop a prediction task called drug mechanism-of-action (MoA) deconvolution and create a tailored knowledge graph (KG), MoA-net. They then design the MoA Retrieval System (MARS), a NeSy approach that leverages logical rules with learned rule weights. MARS uses an interpretable feature alongside domain knowledge to predict drug mechanisms, achieving performance on par with state-of-the-art models while providing aligned model interpretations. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary Neurosymbolic AI combines logic and neural networks for better understanding of biomedical data like drug discovery. The goal is to make sense of how drugs work. There’s no clear way to check if the results are biologically correct, so the authors create a new task and special knowledge graph. They then design an approach that uses rules learned from data to predict drug mechanisms. This approach does as well as others while providing understandable results. |
Keywords
» Artificial intelligence » Knowledge graph
