Powered by OpenAIRE graph
Found an issue? Give us feedback
image/svg+xml art designer at PLoS, modified by Wikipedia users Nina, Beao, JakobVoss, and AnonMoos Open Access logo, converted into svg, designed by PLoS. This version with transparent background. http://commons.wikimedia.org/wiki/File:Open_Access_logo_PLoS_white.svg art designer at PLoS, modified by Wikipedia users Nina, Beao, JakobVoss, and AnonMoos http://www.plos.org/ Estudo Geralarrow_drop_down
image/svg+xml art designer at PLoS, modified by Wikipedia users Nina, Beao, JakobVoss, and AnonMoos Open Access logo, converted into svg, designed by PLoS. This version with transparent background. http://commons.wikimedia.org/wiki/File:Open_Access_logo_PLoS_white.svg art designer at PLoS, modified by Wikipedia users Nina, Beao, JakobVoss, and AnonMoos http://www.plos.org/
Estudo Geral
Master thesis . 2023
Data sources: Estudo Geral
addClaim

Prediction of Drug-Target Binding Affinity: A Regression Approach for IC50Prediction

Authors: Gaspar, Ana Teresa Santos Costa;

Prediction of Drug-Target Binding Affinity: A Regression Approach for IC50Prediction

Abstract

Desde sempre, o ser humano procura combater a doença pela administração de fármacos. A evolução da Ciência, juntamente com a evolução da descoberta de fármacos, conduziu a indústria a focar-se unicamente na produção de fármacos. Atualmente, a indústria farmacêutica procura ativamente propor novos medicamentos, mas esta é uma tarefa que se está a tornar cada vez mais difícil, dado o carácter dispendioso, em termos de custo e tempo, deste processo. Por esta razão, tem- se assistido a um investimento na aplicação de técnicas computacionais para limitar o espaço de procura de potenciais fármacos nas primeiras fases do processo. Nestas etapas, procura-se ainda compreender as interações proteína-ligando para um entendimento detalhado dos mecanismos de ação dos fármacos. Determinar a afinidade de ligação fármaco-proteína é crucial para o reconhecimento de substâncias ativas. Geralmente, é expresso em termos de Kd, Ki ou IC50, sendo que a última é a única que informa acerca da potência do fármaco. Diversos previsores têm sido desenvolvidos, mas apenas consideram Ki ou Kd e interações com cinases. A necessidade por um previsor geral capaz de prever em diferentes tipos de proteínas é evidente. Por esta razão, este trabalho propõe ICARO: um previsor para o cálculo da afinidade de ligação fármaco-proteína, expressa em IC50. O modelo atingiu em teste, R2 de 0.727, rm2 de 0.684, RMSE de 0.700, SCC de 0.855, PCC de 0.855 e CI de 0.837. Quando o ligando era desconhecido, condição S3, os resultados foram de 0.668, 0.634, 0.767, 0.818, 0.819, e 0.816, respetivamente. E, quando a proteína era desconhecida, condição S2, foi obtido um R2 de 0.213, rm2 de 0.246, RMSE de 1.176, SCC de 0.534, PCC de 0.522, e CI de 0.686.

Since ancient times, humans have sought to combat disease through drug administration. As Science has evolved, Drug Discovery (DD) has led industries to focus solely on the production of pharmaceutical products. Currently, the pharmaceutical industry struggles to find new pharmaceutical products; however, this is becoming an increasingly difficult task because it involves an expensive and time-consuming process. Therefore, drug R&D is investing in the application of computational methods for faster narrowing of the number of compounds in the primary stages of DD. The aim of these stages is to understand Protein-Ligand Interactions (PLI), as they provide a detailed comprehension of the drug mechanisms of action. In particular, the determination of Drug-Target Binding Affinity (DTBA) is crucial for hit recognition. DTBA is usually expressed as Kd, Ki, or IC50; however, IC50 is the only measurement that influences drug potency. Several predictors have been developed, but they mainly focus on Ki and Kd and exclusively deal with kinases involving interactions. Therefore, there is a need for a general algorithm capable of predicting the IC50 values of different targets. For this reason, this study proposes ICARO, an approach for DTBA prediction, expressed as IC50. The proposed model ICAROERT attained, in the testing set, a R2 of 0.727, rm2 of 0.684, RMSE of 0.700, SCC of 0.855, PCC of 0.855 and CI of 0.837. When the ligand was unidentified (S3 or cold drug condition), the results were 0.668, 0.634, 0.767, 0.818, 0.819, and 0.816, respectively. When the target was unfamiliar (S2 or the cold target condition), the performance was R2 = 0.213, rm2 = 0.246, RMSE = 1.176, SCC = 0.534, PCC = 0.522, and CI = 0.686.

Dissertação de Mestrado em Biologia Computacional apresentada à Faculdade de Ciências e Tecnologia

Country
Portugal
Related Organizations
Keywords

Machine Learning, Binding Affinity, Protein-Ligand Interactions, Aprendizagem Computacional, ICARO, IC50, Afinidade de Ligação, Interações Proteína-Ligando

  • BIP!
    Impact byBIP!
    selected citations
    These citations are derived from selected sources.
    This is an alternative to the "Influence" indicator, which also reflects the overall/total impact of an article in the research community at large, based on the underlying citation network (diachronically).
    0
    popularity
    This indicator reflects the "current" impact/attention (the "hype") of an article in the research community at large, based on the underlying citation network.
    Average
    influence
    This indicator reflects the overall/total impact of an article in the research community at large, based on the underlying citation network (diachronically).
    Average
    impulse
    This indicator reflects the initial momentum of an article directly after its publication, based on the underlying citation network.
    Average
Powered by OpenAIRE graph
Found an issue? Give us feedback
selected citations
These citations are derived from selected sources.
This is an alternative to the "Influence" indicator, which also reflects the overall/total impact of an article in the research community at large, based on the underlying citation network (diachronically).
BIP!Citations provided by BIP!
popularity
This indicator reflects the "current" impact/attention (the "hype") of an article in the research community at large, based on the underlying citation network.
BIP!Popularity provided by BIP!
influence
This indicator reflects the overall/total impact of an article in the research community at large, based on the underlying citation network (diachronically).
BIP!Influence provided by BIP!
impulse
This indicator reflects the initial momentum of an article directly after its publication, based on the underlying citation network.
BIP!Impulse provided by BIP!
0
Average
Average
Average
Green