AutoMSR: Auto Molecular Structure Representation Learning for Multi-label Metabolic Pathway Prediction

Chen, Jiamin; Gao, Jianliang; Lyu, Tengfei; Oloulade, Babatounde Moctard; Hu, Xiaohua

doi:10.1109/tcbb.2022.3198119

Cited by 5 publications

(3 citation statements)

References 31 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Recently, there has been a lot of attention given to graph neural architecture search (GraphNAS), which is used to automatically find the best neural architecture for graph representation learning task ( Chen et al 2022 , Oloulade et al 2022 , Al-Sabri et al 2023 ). The objective of GraphNAS, given a search space S , and a validation dataset V , is to select the best architecture

that optimizes an evaluation performance

.…”

Section: Introductionmentioning

confidence: 99%

“…The objective of GraphNAS, given a search space S , and a validation dataset V , is to select the best architecture

that optimizes an evaluation performance

. As evaluating the performance of architectures is a time-consuming process and the search space contains millions of architectures, existing frameworks typically stop the search process based on a predefined computation budget ( Chen et al 2022 , Al-Sabri et al 2023 ). There is a disadvantage to this strategy in that it limits the rate of exploration of the search space by the search algorithm.…”

Section: Introductionmentioning

confidence: 99%

“…This reduces the chances of the GraphNAS framework acquiring the optimal architecture. For example, the search space used by Chen et al (2022) contains over

architectures but the search algorithm only explores 2000 architectures. Because the exploration of search algorithms is limited to

, the likelihood of obtaining the global optimal GNN architecture is greatly reduced.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Cancer drug response prediction with surrogate modeling-based graph neural architecture search

et al. 2023

View full text Add to dashboard Cite

Motivation Understanding drug response differences in cancer treatments is one of the most challenging aspects of personalized medicine. Recently, graph neural networks (GNNs) have become state-of-the-art methods in many graph representation learning scenarios in bioinformatics. However, building an optimal handcrafted GNN model for a particular drug sensitivity dataset requires manual design and fine-tuning of the hyperparameters for the GNN model, which is time-consuming and requires expert knowledge. Results In this work, we propose AutoCDRP, a novel framework for automated cancer drug response predictor using graph neural networks (GNNs). Our approach leverages surrogate modeling to efficiently search for the most effective GNN architecture. AutoCDRP uses a surrogate model to predict the performance of GNN architectures sampled from a search space, allowing it to select the optimal architecture based on evaluation performance. Hence, AutoCDRP can efficiently identify the optimal GNN architecture by exploring the performance of all GNN architectures in the search space. Through comprehensive experiments on two benchmark datasets, we demonstrate that the GNN architecture generated by AutoCDRP surpasses state-of-the-art designs. Notably, the optimal GNN architecture identified by AutoCDRP consistently outperforms the best baseline architecture from the first epoch, providing further evidence of its effectiveness. Availability and implementation https://github.com/BeObm/AutoCDRP Supplementary information Supplementary data are available at Bioinformatics online.

show abstract

that optimizes an evaluation performance

.…”

Section: Introductionmentioning

confidence: 99%

“…The objective of GraphNAS, given a search space S , and a validation dataset V , is to select the best architecture

that optimizes an evaluation performance

Section: Introductionmentioning

confidence: 99%

“…This reduces the chances of the GraphNAS framework acquiring the optimal architecture. For example, the search space used by Chen et al (2022) contains over

architectures but the search algorithm only explores 2000 architectures. Because the exploration of search algorithms is limited to

, the likelihood of obtaining the global optimal GNN architecture is greatly reduced.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Cancer drug response prediction with surrogate modeling-based graph neural architecture search

et al. 2023

View full text Add to dashboard Cite

show abstract

A systematic literature review on AutoML for multi-target learning tasks

2023

View full text Add to dashboard Cite

Intelligent Electrochemical Sensors for Precise Identification of Volatile Organic Compounds Enabled by Neural Network Analysis

Li,

Huang,

Witherspoon

et al. 2024

IEEE Sensors J.

View full text Add to dashboard Cite

AutoMSR: Auto Molecular Structure Representation Learning for Multi-label Metabolic Pathway Prediction

Cited by 5 publications

References 31 publications

Cancer drug response prediction with surrogate modeling-based graph neural architecture search

Cancer drug response prediction with surrogate modeling-based graph neural architecture search

A systematic literature review on AutoML for multi-target learning tasks

Intelligent Electrochemical Sensors for Precise Identification of Volatile Organic Compounds Enabled by Neural Network Analysis

Contact Info

Product

Resources

About