Deep Denoising of Raw Biomedical Knowledge Graph From COVID-19 Literature, LitCovid, and Pubtator: Framework Development and Validation

Article Biochemical Research Methods

Pre-training graph neural networks for link prediction in biomedical networks

Yahui Long et al.

Summary: In this article, the authors propose a novel Pre-Training Graph Neural Networks-based framework named PT-GNN to integrate different data sources for link prediction in biomedical networks. The framework effectively learns features for individual nodes from sequence and structure data, and refines the node features by modeling the dependencies among nodes in the network. The authors demonstrate that PT-GNN outperforms state-of-the-art methods for link prediction tasks and that the pre-trained features can improve the performance and reduce training time of existing models.

BIOINFORMATICS (2022)

添加到收藏夹

Article Biochemical Research Methods

Drug-target prediction utilizing heterogeneous bio-linked network embeddings

Nansu Zong et al.

Summary: Researchers conducted a study on modularization for network-based prediction, comparing the performance of various methods through experiments and benchmarking to identify the best prediction strategy from a network and methodological perspective. The results showed that the proposed method outperforms existing methods, demonstrating reliability and scalability in disease-specific prediction tasks.

BRIEFINGS IN BIOINFORMATICS (2021)

添加到收藏夹

Article Computer Science, Artificial Intelligence

Image classification with deep learning in the presence of noisy labels: A survey

Gorkem Algan et al.

Summary: Deep neural networks have made significant progress in image classification systems, but the excessive labeled data required for adequate training poses challenges due to label noise. This paper categorizes methodologies into noise model based and noise model free methods, with the former aiming to estimate noise structure and avoid adverse effects, and the latter focusing on inherently noise robust algorithms using approaches like robust losses.

KNOWLEDGE-BASED SYSTEMS (2021)

添加到收藏夹

Article Biochemistry & Molecular Biology

LitCovid: an open database of COVID-19 literature

Qingyu Chen et al.

Summary: Since the outbreak of the COVID-19 pandemic in 2020, there has been a significant increase in articles related to COVID-19 and SARS-CoV-2. LitCovid, as the first COVID-19-specific literature resource, provides freely accessible curated articles and information, and has been widely used by users globally.

NUCLEIC ACIDS RESEARCH (2021)

添加到收藏夹

Letter Multidisciplinary Sciences