A knowledge graph embedding based approach to predict the adverse drug reactions using a deep neural network

Joshi, Pratik; Masilamani, V.; Mukherjee, Anirban

doi:10.1016/j.jbi.2022.104122

Cited by 27 publications

(10 citation statements)

References 39 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The ReproTox-KG is an initial effort towards integrating knowledge about birth defects, genes, and drugs. Similar efforts have been recently published, including studies that attempted to use graph embedding algorithms to predict missing/novel associations between drugs and diseases [67], for drug repurposing opportunities [68] [69], predicting drug targets [70] [71], adverse events [72], and drug-drug interactions [73]. These are just a few studies in this domain.…”

Section: Discussionmentioning

confidence: 99%

ReproTox-KG: Toxicology Knowledge Graph for Structural Birth Defects

Evangelista

Clarke

Xie

et al. 2022

Preprint

View full text Add to dashboard Cite

Birth defects are functional and structural abnormalities that impact 1 in 33 births in the United States. Birth defects have been attributed to genetic as well as other factors, but for most birth defects there are no known causes. Small molecule drugs, cosmetics, foods, and environmental pollutants may cause birth defects when the mother is exposed to them during pregnancy. These molecules may interfere with the process of normal fetal development. To characterize associations between small molecule compounds and their potential to induce specific birth abnormalities, we gathered knowledge from multiple sources to construct a reproductive toxicity Knowledge Graph (ReproTox-KG) with an initial focus on associations between birth defects, drugs, and genes. Specifically, to construct ReproTox-KG we gathered data from drug/birth-defect associations from co-mentions in published abstracts, gene/birth-defect associations from genetic studies, drug- and preclinical-compound-induced gene expression data, known drug targets, genetic burden scores for all human genes, and placental crossing scores for all small molecules in ReproTox-KG. Using the data stored within ReproTox-KG, we scored 30,000 preclinical small molecules for their potential to induce birth defects. Querying the ReproTox-KG, we identified over 500 birth-defect/gene/drug cliques that can be used to explain molecular mechanisms for drug-induced birth defects. The ReproTox-KG is provided as curated tables and via a web-based user interface that can enable users to explore the associations between birth defects, approved and preclinical drugs, and human genes.

show abstract

Section: Discussionmentioning

confidence: 99%

ReproTox-KG: Toxicology Knowledge Graph for Structural Birth Defects

Evangelista

Clarke

Xie

et al. 2022

Preprint

View full text Add to dashboard Cite

show abstract

“…There are several methods for predicting ADRs, mainly including multi-label methods [4,5] and methods based on drug-ADR association [6][7][8] . The former uses the characteristics of the drug itself to predict the probability of relevant ADRs and the latter uses the association information between drugs and ADRs to model and calculate the probability of occurrence of drug-ADR pairs.…”

Section: Introductionmentioning

confidence: 99%

A novel pseudo-labeling framework based on matrix integration for adverse drug reaction prediction

Chen,

Han,

et al. 2024

Preprint

View full text Add to dashboard Cite

Background Adverse drug reactions (ADRs) are one of the global public health events that seriously endanger human life and cause high economic burdens. Therefore, predicting the possibility of their occurrence and taking early and effective response measures are of great significance. The construction of a correlation matrix between drugs and their adverse reactions, followed by effective correlation data mining, is one of the current strategies to predict ADRs using accessible public data. Since the number of known ADRs in real-world data is far less than the number of their unknown counterparts, the drug-ADR association matrix is very sparse, which greatly affects the classification performance of machine learning methods. Results To effectively solve the problem of sparsity, we propose a novel pseudo-labeled data augmentation framework that mines potential unknown drug-ADR pairs by integrating multiple weighted matrix factorization (MF) models and treating them as pseudo-labeled drug-ADR pairs. Pseudo-labeled data is added to the training set and the MF model is fine-tuned to improve the classification performance. Conclusions This paper reproduces the baselines under the same experimental conditions to evaluate the performance of the proposed method on sparse data from the Side Effect Resource (SIDER) database. Experimental results showed that our method outperformed other baselines in the Area Under Precision-Recall (AUPR) and F1 scores, and still maintained the best performance in sparser scenarios. Furthermore, we conducted a case study and the results showed that our proposed framework efficiently predicted ADRs in the real world.

show abstract

“…For that purpose, similarity measures [ 29 ] and representative KG embeddings of chemical drug structures via neural networks [ 30 ] have been used in prediction approaches. Other authors proposed network representation learning techniques and graph regularized matrix factorization for predicting ADEs of individual drugs [ 31 , 32 ]. Also, ensembles of several learning techniques have been tested [ 33 ].…”

Section: Introductionmentioning

confidence: 99%

MultiGML: Multimodal graph machine learning for prediction of adverse drug events

Krix,

DeLong,

Madan

et al. 2023

Heliyon

View full text Add to dashboard Cite

A knowledge graph embedding based approach to predict the adverse drug reactions using a deep neural network

Cited by 27 publications

References 39 publications

ReproTox-KG: Toxicology Knowledge Graph for Structural Birth Defects

ReproTox-KG: Toxicology Knowledge Graph for Structural Birth Defects

A novel pseudo-labeling framework based on matrix integration for adverse drug reaction prediction

MultiGML: Multimodal graph machine learning for prediction of adverse drug events

Contact Info

Product

Resources

About