▲ Schematic diagram of a computer methodology for predicting metabolites and metabolic pathways associated with cancer somatic mutations (Photo = Provided by KAIST)
[메디컬투데이=이재혁 기자] Unlike normal cells, cancer undergoes metabolic reactions caused by abnormal accumulation within cells, and these cancer metabolic reactions are being studied in a variety of ways for the purpose of cancer treatment and diagnosis.
Accordingly, KAIST researchers succeeded in building metabolic models for 1,043 cancer patients corresponding to 24 types of cancer using computers.
At KAIST, the research team of Professor Hyunwook Kim and Distinguished Professor Sangyeop Lee of the Department of Biochemical Engineering, through joint research with the research team of Professor Youngil Ko, Professor Hongseok Yoon, and Professor Changwook Jeong at Seoul National University Hospital, developed a computer methodology to predict new metabolites and metabolic pathways associated with cancer somatic gene mutations. It was announced on the 18th that it had been developed.
Recently, the discovery of cancer-causing metabolites (oncometabolites) and new drugs targeting them have been receiving attention with the approval of the U.S. Food and Drug Administration (FDA), including ‘Tipsovo (ingredient name: ivosidenib), which is being used as a treatment for acute myeloid leukemia. and the drug ‘Eidhyfa (ingredient name: enasidenib)’.
However, research on cancer metabolism and discovery of new cancer-causing metabolites requires methodologies such as metabolomics, and applying them to large-scale patient samples takes considerable time and cost. For this reason, although many genetic mutations associated with cancer have been identified, only a few corresponding cancer-causing metabolites are known.
Professor Hyunwook Kim’s joint research team integrated the transcriptome data of cancer patients released by the International Cancer Research Consortium into a genome-level metabolic model that can predict cell metabolism information, and created a metabolic model for 1,043 cancer patients corresponding to 24 cancer types. was successfully built.
The joint research team used 1,043 cancer patient-specific metabolic models and cancer somatic mutation data from the same patients to develop a computer methodology consisting of the following four steps.
In the first step, a cancer patient-specific metabolic model is simulated to predict the activity of all metabolites for each patient. In the second step, pairs of specific gene mutations that cause significant differences in the activity of previously predicted metabolites are selected. In the third step, metabolites linked to specific gene mutations are further selected to identify metabolic pathways significantly associated with them. As a final step, the ‘gene-metabolite-metabolic pathway’ combination is completed and derived as a computer methodology result.
Dr. Garyeong Lee (currently a postdoctoral researcher at Dana-Farber Cancer Center and Harvard Medical School) and Dr. Lee Sang-mi (currently a postdoctoral researcher at Harvard Medical School), co-first authors of this paper, said, “The methodology developed in this study “It can be easily applied to other cancer types based on transcriptome data, and it is significant in that it is the first computer methodology that can systematically predict how genetic mutations cause changes in cell metabolism through metabolic pathways,” he said.
In addition, KAIST Professor Kim Hyun-wook emphasized, “The results of this joint research can be used as important reference material in future research on cancer metabolism and cancer-causing metabolites.”
Meanwhile, this paper is published by BioMed Central and published in Genome Biology (JCR, within the top 5%), a leading international journal in the fields of biotechnology and genetics.
[ⓒ 메디컬투데이. 무단전재-재배포 금지]
2024-03-17 23:00:25
#KAISTSNUH #joint #research #team #predicts #cancercausing #substances #computer
