Summary of Structure-based Drug Design Benchmark: Do 3d Methods Really Dominate?, by Kangyu Zheng et al.
Structure-based Drug Design Benchmark: Do 3D Methods Really Dominate?
by Kangyu Zheng, Yingzhou Lu, Zaixi Zhang, Zhongwei Wan, Yao Ma, Marinka Zitnik, Tianfan Fu
First submitted to arxiv on: 4 Jun 2024
Categories
- Main: Machine Learning (cs.LG)
- Secondary: Artificial Intelligence (cs.AI); Quantitative Methods (q-bio.QM)
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 The paper proposes a benchmark to evaluate the performance of 16 models from different algorithmic foundations in structure-based drug design (SBDD). The authors compare and contrast three main types of algorithms: search-based, deep generative models, and reinforcement learning. They assess the pharmaceutical properties of generated molecules and their docking affinities with target proteins. The results show that 1D/2D ligand-centric methods can be competitive with 3D-based methods, and AutoGrow4 dominates SBDD in terms of optimization ability. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary This paper compares different ways to design new medicines using computer algorithms. It looks at three main types of algorithms: search-based, deep learning models, and reinforcement learning. The authors test these algorithms by seeing how well they work together with a target protein. They find that some simpler algorithms can be just as good as more complex ones, and one algorithm called AutoGrow4 is particularly good at finding new medicines. |
Keywords
» Artificial intelligence » Deep learning » Optimization » Reinforcement learning
