Summary of Molecular Topological Profile (moltop) — Simple and Strong Baseline For Molecular Graph Classification, by Jakub Adamczyk et al.
Molecular Topological Profile (MOLTOP) – Simple and Strong Baseline for Molecular Graph Classification
by Jakub Adamczyk, Wojciech Czech
First submitted to arxiv on: 16 Jul 2024
Categories
- Main: Machine Learning (cs.LG)
- Secondary: 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 This paper challenges the effectiveness of topological descriptors in molecular graph classification by designing a simple yet strong baseline approach. The authors demonstrate that a combination of histogram aggregation and one-hot encoding with a Random Forest classifier can establish a competitive baseline for Graph Neural Networks (GNNs). They introduce the Molecular Topological Profile (MOLTOP) algorithm, which integrates Edge Betweenness Centrality, Adjusted Rand Index, and SCAN Structural Similarity score. This approach is shown to be remarkably competitive compared to modern GNNs, being simple, fast, low-variance, and hyperparameter-free. The authors rigorously test their approach on MoleculeNet datasets using the fair evaluation protocol provided by Open Graph Benchmark and also demonstrate out-of-domain generation capabilities on peptide classification task from Long Range Graph Benchmark. Across eleven benchmark datasets, the evaluations reveal MOLTOP’s strong discriminative capabilities, surpassing the 1-WL test and even 3-WL test for some classes of graphs. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary This paper looks at how well topological descriptors work in classifying molecules. The authors create a simple approach that uses histogram aggregation and one-hot encoding to classify molecules using a Random Forest classifier. They call this approach the Molecular Topological Profile (MOLTOP). MOLTOP is compared to more advanced neural networks, and surprisingly it performs just as well! This is because MOLTOP is easy to use, fast, and doesn’t require adjusting many parameters. The authors test their approach on lots of different datasets and show that it works really well. |
Keywords
» Artificial intelligence » Classification » Hyperparameter » One hot » Random forest
