University of Tehran Researchers Use AI to Analyze Biological Data, Pathology Images

A group of researchers from the faculties of engineering at the University of Tehran, led by Ali Massoudinejad, a professor of systems biology and bioinformatics at the Faculty of Engineering Sciences, in collaboration with Behnaz Hajimollahosseini, a PhD student, Ahmadreza Iranpour, a student at Utrecht University in the Netherlands, and Sevda Imani, a student at Pázsony University in Hungary, and other researchers at the Systems Biology and Bioinformatics Laboratory, succeeded in conducting two innovative scientific studies in the field of deep learning and personalized medicine.

“In the first study, we categorized the methods into four main groups of deep learning-based methods like Generative Adversarial Network (GAN) and autoencoder, traditional methods like histogram matching, hybrid models, and a new method based on signal processing, and showed that each of these approaches has its own advantages and limitations,” said Massoudinejad.

“The results of our study emphasize the importance of preserving biological information in the normalization process and its role in increasing the accuracy of computer-aided diagnosis systems,” he added.

Elaborating on the second study, Massoudinejad said, “In this study, the focus was on analyzing multi-omics data and applying deep learning models to predict survival in patients with endometrial cancer.”

“In this study, gene expression, DNA methylation, and proteome data from the TCGA-UCEC project were examined, and a new autoencoder with a dedicated cost function was designed to better identify the complex nonlinear relationships between biological features and survival rates,” he added.

“The results showed that this approach extracts survival-related information more accurately than conventional methods and led to the identification of key molecular pathways, such as the vitamin D pathway and the galanin receptor, that are associated with patient prognosis,” Massoudinejad said.

In a relevant development in 2024, caretaker of the National Center for Genetic and Biological Resources of Iran Jafar Razeqi said that the country has made a special network to protect its genetic and biological reserves and resources.

“Due to different ecosystems and climatic conditions, Iran has different genetic and biological resources and in many cases unique, and it can be said that no country in the world has this climate and distribution in the field of genetic and biological resources,” said Razeqi.

“In 15 years of its activity, the National Center for Genetic and Biological Resources of Iran has been able to set up biobanks in four areas of plants, human and animal cells, microorganisms and DNA and genomic data while collecting, identifying, producing and storing them according to its missions,” he added.

“We have created a network to protect the genetic and biological resources of Iran,” Razeqi said.

4155/v