Summary of Predicting Drug Effects From High-dimensional, Asymmetric Drug Datasets by Using Graph Neural Networks: a Comprehensive Analysis Of Multitarget Drug Effect Prediction, By Avishek Bose and Guojing Cong
Predicting Drug Effects from High-Dimensional, Asymmetric Drug Datasets by Using Graph Neural Networks: A Comprehensive Analysis of Multitarget Drug Effect Prediction
by Avishek Bose, Guojing Cong
First submitted to arxiv on: 11 Oct 2024
Categories
- Main: Machine Learning (cs.LG)
- Secondary: None
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 Graph neural networks (GNNs) have been successful in predicting drug effects from molecular graphs. However, current GNN models struggle when faced with high-dimensional, asymmetrically co-occurrent drug effects with complex correlations. To address this challenge, we propose multitarget prediction problems that predict multiple drug effects simultaneously. We developed standard and hybrid GNNs for multiregression (continuous values) and multilabel classification (categorical values). The latter is particularly challenging due to sparse target data and asymmetric label co-occurrence. To improve performance, we introduce a new data oversampling technique, which enhances the data imbalance ratio while preserving dataset integrity. Our best-performing hybrid GNN model achieves state-of-the-art results in precision, recall, and F1 score on our proposed oversampled datasets. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary Scientists have been trying to use computers to predict how different medicines work together. They’ve had some success, but it’s hard when they’re dealing with lots of complex information. Instead of predicting one medicine at a time, researchers propose predicting multiple medicines at once. To make this happen, they developed new computer models that can handle both continuous and categorical data. The challenge is that the data is very sparse and the relationships between different medicines are tricky to understand. To solve this problem, the team came up with a new way to prepare the data, which helps their model perform better. |
Keywords
» Artificial intelligence » Classification » F1 score » Gnn » Precision » Recall
