IT Blog


Researchers from the University of Coimbra create model to develop cancer drugs faster and less expensive – Fall River Herald News

COIMBRA – A team of researchers from the University of Coimbra (UC) has designed an innovative computational model that develops new drugs for cancer treatment faster and less expensive.
The model results from a collaboration between the Faculty of Science and Technology (FCTUC) and the Faculty of Pharmacy (FFUC), using Artificial Intelligence through computational methods “that can generate pharmacologically interesting compounds in a faster, automated way,” the UC said in a statement sent to Lusa news agency on Monday.
“Considering that the discovery of a drug is an extremely complex, lengthy and costly process, this work aimed to shorten the initial stages of drug development,” according to the statement.
The team from the Computer Engineering Department at FCTUC used Machine Learning techniques, namely Deep Learning – a method that uses artificial neural networks, which create intelligent models “by mimicking the learning capacity of biological models.”
This way, “they can identify patterns embedded in data sets and, from there, it is possible to obtain models that generate new molecular structures that predict biological properties of interest,” explained Tiago Oliveira Pereira, first author of the study which is part of his PhD.
The researchers also used the so-called Reinforcement Learning, which optimizes the generative model while exploring the existing chemical space.
“As the model generates new molecules, it receives a reward, which will be higher or lower depending on the state of optimization of the compounds’ properties. Thus, throughout this optimization process, the compound generator will learn to identify the regions of the chemical space that allow it to obtain greater reward and better compounds,” Tiago Oliveira Pereira said.
According to the authors, the model is innovative because “it combines chemical information, through the compounds, and biological, through the information of the gene expression, to find promising molecules in the inhibition of the receptor and that do not cause undesirable effects to the biological system.”
With the collaboration of the laboratory of Professor Jorge Salvador of FFUC, it was possible to apply the model in a case study for the generation of compounds capable of inhibiting the protein USP7 (Ubiquitin specific protease 7).
The model was applied in a case study to generate compounds capable of inhibiting the USP7 protein (Ubiquitin specific protease 7), which plays a key role “in the progression of various types of cancer and is currently seen as an essential receptor for drug development.”
The UC stressed that the results obtained in the experiments are encouraging, with the model having demonstrated a high capacity to generate potential USP7 inhibitor molecules.
“More than 90% of the molecules contained physical, chemical and biological properties essential for interaction with the receptor. In addition, we found that some compounds generated by the model have similarities with anti-cancer drugs in terms of their active groups, which validates the approach implemented,” said Tiago Oliveira Pereira.
Despite being validated with breast cancer data, the new computational model may be applied to “several contexts in which gene expression data associated with disease progression may be obtained.”
The researcher also said that the next steps of the investigation would focus on improving the implemented architecture and the “definition of a set of validation methods to filter the molecules obtained and, depending on the results, move forward to the synthesis of the best compounds.”
The Foundation for Science and Technology, the Central Administration Investment and Development Expenditure Programme and European funds co-financed the study through the D4-Deep Drug Discovery and Deployment project.