Modeling Pharmacological Effects with Multi-Relation Unsupervised Graph Embedding
Autor(a) principal: | |
---|---|
Data de Publicação: | 2020 |
Tipo de documento: | Dissertação |
Idioma: | eng |
Título da fonte: | Repositório Institucional da UFMG |
Texto Completo: | http://hdl.handle.net/1843/42300 |
Resumo: | Drug repositioning (aka repurposing) can be defined as renewing failed drugs (proved safety but failed to show efficacy for their primary indication) and expanding successful ones by developing new therapeutic uses that are beyond their original uses or initial approved indications. Repositioned drugs account for approximately 30\% of the US Food and Drug Administration (FDA) approved drugs in recent years. A repositioned drug uses de-risked compounds, going directly to preclinical testing and clinical trials, thus providing inexpensive alternatives to the costly pipeline associated with the development of new drugs. A pharmacological effect of a drug on cells, organs and systems refers to the specific biochemical interaction produced by a drug substance, which is called its mechanism of action. There are several approaches for novel repositioning opportunities identification, such as signature matching, molecular docking and genetic association in literature. In this work, we present a novel method based on a multi-relation unsupervised graph embedding model that learns latent representations for drugs (mechanisms of action) and diseases so that the distance between these representations reveals repositioning opportunities. Once representations for drugs and diseases are obtained we learn the likelihood of new links (that is, new indications) between drugs and diseases. Known drug indications are used for learning a model that predicts potential indications. Compared with existing unsupervised graph embedding methods our method shows superior prediction performance in terms of area under the ROC curve, and we present examples of repositioning opportunities found on recent biomedical literature that were also predicted by our method. |
id |
UFMG_6abf5bbd87f4b0d6d2c65e97d6b8452e |
---|---|
oai_identifier_str |
oai:repositorio.ufmg.br:1843/42300 |
network_acronym_str |
UFMG |
network_name_str |
Repositório Institucional da UFMG |
repository_id_str |
|
spelling |
Adriano Alonso Velosohttp://lattes.cnpq.br/9973021912226739Nivio ZivianiRaquel Cardoso de Melo MinardiDeborah Schechtmanhttp://lattes.cnpq.br/1641521413108988Dehua Chen2022-06-07T00:22:17Z2022-06-07T00:22:17Z2020-03-30http://hdl.handle.net/1843/42300Drug repositioning (aka repurposing) can be defined as renewing failed drugs (proved safety but failed to show efficacy for their primary indication) and expanding successful ones by developing new therapeutic uses that are beyond their original uses or initial approved indications. Repositioned drugs account for approximately 30\% of the US Food and Drug Administration (FDA) approved drugs in recent years. A repositioned drug uses de-risked compounds, going directly to preclinical testing and clinical trials, thus providing inexpensive alternatives to the costly pipeline associated with the development of new drugs. A pharmacological effect of a drug on cells, organs and systems refers to the specific biochemical interaction produced by a drug substance, which is called its mechanism of action. There are several approaches for novel repositioning opportunities identification, such as signature matching, molecular docking and genetic association in literature. In this work, we present a novel method based on a multi-relation unsupervised graph embedding model that learns latent representations for drugs (mechanisms of action) and diseases so that the distance between these representations reveals repositioning opportunities. Once representations for drugs and diseases are obtained we learn the likelihood of new links (that is, new indications) between drugs and diseases. Known drug indications are used for learning a model that predicts potential indications. Compared with existing unsupervised graph embedding methods our method shows superior prediction performance in terms of area under the ROC curve, and we present examples of repositioning opportunities found on recent biomedical literature that were also predicted by our method.O reposicionamento de medicamentos (também conhecido como reaproveitamento) pode ser definido como a renovação de medicamentos não aprovados (com uso seguro comprovado, mas não demonstrou eficácia na indicação primária) e a expansão de uso dos medicamentos aprovados, desenvolvendo novos usos terapêuticos, que estão além dos seus usos originais inicialmente aprovados. Os medicamentos reposicionados representam aproximadamente 30% dos medicamentos aprovados pela Food and Drug Administration (FDA) dos EUA nos últimos anos. Um fármaco reposicionado usa compostos de menor risco, podendo ir diretamente para testes pré-clínicos e ensaios clínicos, fornecendo assim alternativas mais baratas comparando ao pipeline caro do desenvolvimento de novos fármacos. Um efeito farmacológico de um medicamento nas células, órgãos e sistemas refere-se à interação bioquímica específica produzida por um medicamento, também chamado como mecanismo de ação. Existem várias abordagens para a identificação de novas oportunidades de reposicionamento, como correspondência de assinatura, docagem molecular (acoplamento molecular, or ancoragem molecular) e associação genética na literatura. Neste trabalho, apresentamos um novo método baseado em um modelo de representações não supervisionadas de grafos multi-relacionais que aprende representações latentes de medicamentos (mecanismo de ação) e doenças, de modo que a distância entre essas representações revele oportunidades de reposicionamento. Uma vez obtidas representações de medicamentos e doenças, aprendemos a predizer a probabilidade de novas indicações entre medicamentos e doenças. As indicações conhecidas de medicamentos são usadas para aprender um modelo que prediz potenciais novas indicações de medicamentos. Comparado com os métodos existentes de representações não supervisionadas de grafos, nosso método mostra desempenho superior em termos de área abaixo da curva ROC (area under the ROC curve ). Também apresentamos exemplos de oportunidades de reposicionamento encontradas na literatura biomédica recente que também foram previstas pelo nosso método.engUniversidade Federal de Minas GeraisPrograma de Pós-Graduação em Ciência da ComputaçãoUFMGBrasilICX - DEPARTAMENTO DE CIÊNCIA DA COMPUTAÇÃOComputação – TesesAprendizado do computador – TesesAprendizado de Representações – TesesDrug RepositioningGraph EmbeddingNode EmbeddingDrug RepurposingModeling Pharmacological Effects with Multi-Relation Unsupervised Graph Embeddinginfo:eu-repo/semantics/publishedVersioninfo:eu-repo/semantics/masterThesisinfo:eu-repo/semantics/openAccessreponame:Repositório Institucional da UFMGinstname:Universidade Federal de Minas Gerais (UFMG)instacron:UFMGORIGINALchen_master_thesis.pdfchen_master_thesis.pdfcorreções feitasapplication/pdf394726https://repositorio.ufmg.br/bitstream/1843/42300/3/chen_master_thesis.pdf058a9e93094afa086820e9057e43f306MD53LICENSElicense.txtlicense.txttext/plain; charset=utf-82118https://repositorio.ufmg.br/bitstream/1843/42300/4/license.txtcda590c95a0b51b4d15f60c9642ca272MD541843/423002022-06-06 21:22:17.719oai:repositorio.ufmg.br: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ório de PublicaçõesPUBhttps://repositorio.ufmg.br/oaiopendoar:2022-06-07T00:22:17Repositório Institucional da UFMG - Universidade Federal de Minas Gerais (UFMG)false |
dc.title.pt_BR.fl_str_mv |
Modeling Pharmacological Effects with Multi-Relation Unsupervised Graph Embedding |
title |
Modeling Pharmacological Effects with Multi-Relation Unsupervised Graph Embedding |
spellingShingle |
Modeling Pharmacological Effects with Multi-Relation Unsupervised Graph Embedding Dehua Chen Drug Repositioning Graph Embedding Node Embedding Drug Repurposing Computação – Teses Aprendizado do computador – Teses Aprendizado de Representações – Teses |
title_short |
Modeling Pharmacological Effects with Multi-Relation Unsupervised Graph Embedding |
title_full |
Modeling Pharmacological Effects with Multi-Relation Unsupervised Graph Embedding |
title_fullStr |
Modeling Pharmacological Effects with Multi-Relation Unsupervised Graph Embedding |
title_full_unstemmed |
Modeling Pharmacological Effects with Multi-Relation Unsupervised Graph Embedding |
title_sort |
Modeling Pharmacological Effects with Multi-Relation Unsupervised Graph Embedding |
author |
Dehua Chen |
author_facet |
Dehua Chen |
author_role |
author |
dc.contributor.advisor1.fl_str_mv |
Adriano Alonso Veloso |
dc.contributor.advisor1Lattes.fl_str_mv |
http://lattes.cnpq.br/9973021912226739 |
dc.contributor.advisor-co1.fl_str_mv |
Nivio Ziviani |
dc.contributor.referee1.fl_str_mv |
Raquel Cardoso de Melo Minardi |
dc.contributor.referee2.fl_str_mv |
Deborah Schechtman |
dc.contributor.authorLattes.fl_str_mv |
http://lattes.cnpq.br/1641521413108988 |
dc.contributor.author.fl_str_mv |
Dehua Chen |
contributor_str_mv |
Adriano Alonso Veloso Nivio Ziviani Raquel Cardoso de Melo Minardi Deborah Schechtman |
dc.subject.por.fl_str_mv |
Drug Repositioning Graph Embedding Node Embedding Drug Repurposing |
topic |
Drug Repositioning Graph Embedding Node Embedding Drug Repurposing Computação – Teses Aprendizado do computador – Teses Aprendizado de Representações – Teses |
dc.subject.other.pt_BR.fl_str_mv |
Computação – Teses Aprendizado do computador – Teses Aprendizado de Representações – Teses |
description |
Drug repositioning (aka repurposing) can be defined as renewing failed drugs (proved safety but failed to show efficacy for their primary indication) and expanding successful ones by developing new therapeutic uses that are beyond their original uses or initial approved indications. Repositioned drugs account for approximately 30\% of the US Food and Drug Administration (FDA) approved drugs in recent years. A repositioned drug uses de-risked compounds, going directly to preclinical testing and clinical trials, thus providing inexpensive alternatives to the costly pipeline associated with the development of new drugs. A pharmacological effect of a drug on cells, organs and systems refers to the specific biochemical interaction produced by a drug substance, which is called its mechanism of action. There are several approaches for novel repositioning opportunities identification, such as signature matching, molecular docking and genetic association in literature. In this work, we present a novel method based on a multi-relation unsupervised graph embedding model that learns latent representations for drugs (mechanisms of action) and diseases so that the distance between these representations reveals repositioning opportunities. Once representations for drugs and diseases are obtained we learn the likelihood of new links (that is, new indications) between drugs and diseases. Known drug indications are used for learning a model that predicts potential indications. Compared with existing unsupervised graph embedding methods our method shows superior prediction performance in terms of area under the ROC curve, and we present examples of repositioning opportunities found on recent biomedical literature that were also predicted by our method. |
publishDate |
2020 |
dc.date.issued.fl_str_mv |
2020-03-30 |
dc.date.accessioned.fl_str_mv |
2022-06-07T00:22:17Z |
dc.date.available.fl_str_mv |
2022-06-07T00:22:17Z |
dc.type.status.fl_str_mv |
info:eu-repo/semantics/publishedVersion |
dc.type.driver.fl_str_mv |
info:eu-repo/semantics/masterThesis |
format |
masterThesis |
status_str |
publishedVersion |
dc.identifier.uri.fl_str_mv |
http://hdl.handle.net/1843/42300 |
url |
http://hdl.handle.net/1843/42300 |
dc.language.iso.fl_str_mv |
eng |
language |
eng |
dc.rights.driver.fl_str_mv |
info:eu-repo/semantics/openAccess |
eu_rights_str_mv |
openAccess |
dc.publisher.none.fl_str_mv |
Universidade Federal de Minas Gerais |
dc.publisher.program.fl_str_mv |
Programa de Pós-Graduação em Ciência da Computação |
dc.publisher.initials.fl_str_mv |
UFMG |
dc.publisher.country.fl_str_mv |
Brasil |
dc.publisher.department.fl_str_mv |
ICX - DEPARTAMENTO DE CIÊNCIA DA COMPUTAÇÃO |
publisher.none.fl_str_mv |
Universidade Federal de Minas Gerais |
dc.source.none.fl_str_mv |
reponame:Repositório Institucional da UFMG instname:Universidade Federal de Minas Gerais (UFMG) instacron:UFMG |
instname_str |
Universidade Federal de Minas Gerais (UFMG) |
instacron_str |
UFMG |
institution |
UFMG |
reponame_str |
Repositório Institucional da UFMG |
collection |
Repositório Institucional da UFMG |
bitstream.url.fl_str_mv |
https://repositorio.ufmg.br/bitstream/1843/42300/3/chen_master_thesis.pdf https://repositorio.ufmg.br/bitstream/1843/42300/4/license.txt |
bitstream.checksum.fl_str_mv |
058a9e93094afa086820e9057e43f306 cda590c95a0b51b4d15f60c9642ca272 |
bitstream.checksumAlgorithm.fl_str_mv |
MD5 MD5 |
repository.name.fl_str_mv |
Repositório Institucional da UFMG - Universidade Federal de Minas Gerais (UFMG) |
repository.mail.fl_str_mv |
|
_version_ |
1803589501684023296 |