Application of Word Embedding to Drug Repositioning

Ngo, Duc Luu; Yamamoto, Naoki; Vu, Tran Anh; Nguyen, Nhat Minh; Phan, Dau; Lumbanraja, Favorisen Rosyking; Kubo, Mamoru; Satou, Kenji

doi:10.4236/jbise.2016.91002

Cited by 20 publications

(9 citation statements)

References 10 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Word embeddings have previously been used for named entity recognition (Habibi et al, 2017) and identifying adverse drug reactions from social media (Nikfarjam et al, 2015). Recently, they have also been included along with a variety of other data sets in models for drug repurposing in different human diseases (Ngo et al, 2016;Manchanda and Anand, 2017). GloVe has been applied to corpora as large as 840 billion words (Pennington et al, 2014) and has been used on tasks as diverse as annotating videos from free-text descriptions (Hendricks et al, 2017) to identifying implicit human bias/stereotyping (Greenwald, 2017).…”

Section: Discussionmentioning

confidence: 99%

Drug Repurposing Prediction for Immune-Mediated Cutaneous Diseases using a Word-Embedding–Based Machine Learning Approach

Patrick

Raja

Miller

et al. 2019

Journal of Investigative Dermatology

View full text Add to dashboard Cite

Immune-mediated diseases affect more than 20% of the population, and many autoimmune diseases affect the skin. Drug repurposing (or repositioning) is a cost-effective approach for finding drugs that can be used to treat diseases for which they are currently not prescribed. We implemented an efficient bioinformatics approach using word embedding to summarize drug information from more than 20 million articles and applied machine learning to model the drug-disease relationship. We trained our drug repurposing approach separately on nine cutaneous diseases (including psoriasis, atopic dermatitis, and alopecia areata) and eight other immunemediated diseases and obtained a mean area under the receiver operating characteristic of 0.93 in crossvalidation. Focusing in particular on psoriasis, a chronic inflammatory condition of skin that affects more than 100 million people worldwide, we were able to confirm drugs that are known to be effective for psoriasis and to identify potential candidates used to treat other diseases. Furthermore, the targets of drug candidates predicted by our approach were significantly enriched among genes differentially expressed in psoriatic lesional skin from a large-scale RNA sequencing cohort. Although our algorithm cannot be used to determine clinical efficacy, our work provides an approach for suggesting drugs for repurposing to immune-mediated cutaneous diseases.

show abstract

Section: Discussionmentioning

confidence: 99%

Drug Repurposing Prediction for Immune-Mediated Cutaneous Diseases using a Word-Embedding–Based Machine Learning Approach

Patrick

Raja

Miller

et al. 2019

Journal of Investigative Dermatology

View full text Add to dashboard Cite

show abstract

“…Listed in Table 3 are the 24 predicted drugs, with their names and whether or not they had been reported in PubMed, ClinicalTrials.gov, or both. Nineteen (19) of them have PubMed support, meaning they were studied in various IBC models according to PubMed publications. Eleven (11) of the 24 drugs with different mechanisms of action had been tested in clinical trials with varying degrees of clinical benefits for IBC [30][31][32][33][34][35][36][37].…”

Section: Literature Validationmentioning

confidence: 99%

“…The words are positioned in the vector space such that those with common linguistic context are found in close proximity. It has been employed for drug repurposing in a couple of cases [18,19].…”

Section: Introductionmentioning

confidence: 99%

Literature-Wide Association Studies (LWAS) for a Rare Disease: Drug Repurposing for Inflammatory Breast Cancer

Jin

Dong

et al. 2020

Molecules

View full text Add to dashboard Cite

Drug repurposing is an effective means for rapid drug discovery. The aim of this study was to develop and validate a computational methodology based on Literature-Wide Association Studies (LWAS) of PubMed to repurpose existing drugs for a rare inflammatory breast cancer (IBC). We have developed a methodology that conducted LWAS based on the text mining technology Word2Vec. 3.80 million “cancer”-related PubMed abstracts were processed as the corpus for Word2Vec to derive vector representation of biological concepts. These vectors for drugs and diseases served as the foundation for creating similarity maps of drugs and diseases, respectively, which were then employed to find potential therapy for IBC. Three hundred and thirty-six (336) known drugs and three hundred and seventy (370) diseases were expressed as vectors in this study. Nine hundred and seventy (970) previously known drug-disease association pairs among these drugs and diseases were used as the reference set. Based on the hypothesis that similar drugs can be used against similar diseases, we have identified 18 diseases similar to IBC, with 24 corresponding known drugs proposed to be the repurposing therapy for IBC. The literature search confirmed most known drugs tested for IBC, with four of them being novel candidates. We conclude that LWAS based on the Word2Vec technology is a novel approach to drug repurposing especially useful for rare diseases.

show abstract

“…Many studies have employed simple classifiers to extract information from texts. For example, Ngo et al [ 25 ] employed a classification method on a set of features based on distributed representations to predict drug-disease relations in cancer treatment. Rastegar-Mojarad et al [ 26 ] used machine learning methods to identify disease names from user reviews for about top 180 most frequently searched medications on the WebMD forum.…”

Section: Related Workmentioning

confidence: 99%

Combination of Deep Recurrent Neural Networks and Conditional Random Fields for Extracting Adverse Drug Reactions from User Reviews

Tutubalina

Nikolenko

2017

Journal of Healthcare Engineering

View full text Add to dashboard Cite

Adverse drug reactions (ADRs) are an essential part of the analysis of drug use, measuring drug use benefits, and making policy decisions. Traditional channels for identifying ADRs are reliable but very slow and only produce a small amount of data. Text reviews, either on specialized web sites or in general-purpose social networks, may lead to a data source of unprecedented size, but identifying ADRs in free-form text is a challenging natural language processing problem. In this work, we propose a novel model for this problem, uniting recurrent neural architectures and conditional random fields. We evaluate our model with a comprehensive experimental study, showing improvements over state-of-the-art methods of ADR extraction.

show abstract

Application of Word Embedding to Drug Repositioning

Cited by 20 publications

References 10 publications

Drug Repurposing Prediction for Immune-Mediated Cutaneous Diseases using a Word-Embedding–Based Machine Learning Approach

Drug Repurposing Prediction for Immune-Mediated Cutaneous Diseases using a Word-Embedding–Based Machine Learning Approach

Literature-Wide Association Studies (LWAS) for a Rare Disease: Drug Repurposing for Inflammatory Breast Cancer

Combination of Deep Recurrent Neural Networks and Conditional Random Fields for Extracting Adverse Drug Reactions from User Reviews

Contact Info

Product

Resources

About