Scientific Reports (Mar 2023)

Artificial intelligence reveals dysregulation of osteosarcoma and cuproptosis-related biomarkers, PDHA1, CDKN2A and neutrophils

  • Jie Jiang,
  • Xinli Zhan,
  • Jianxun Wei,
  • Qie Fan,
  • Haowen Li,
  • Hao Li,
  • Shuzhen Li,
  • Yong Zhao,
  • Guodong Yin,
  • Lin Tang,
  • Yongxiang Wu,
  • Mindong Lan,
  • Yijue Qin,
  • Quan Guo,
  • Weicheng Xu,
  • Ling Lu,
  • Yanwei Yang,
  • Yitian Zhang,
  • Haishun Qu

DOI
https://doi.org/10.1038/s41598-023-32195-2
Journal volume & issue
Vol. 13, no. 1
pp. 1 – 16

Abstract

Read online

Abstract At present, the impact of cuproptosis-related genes in the study of osteosarcoma is largely unknown. Genome-wide data of osteosarcoma and controls were downloaded from 3 different databases, and specific diagnostic models associated with cuproptosis in osteosarcoma were constructed by support vector machines with artificial intelligence, random forest trees and LASSO regression. Differential analysis of immune cell infiltration was examined using routine blood data from 25,665 cases. Differential expression was examined using immunohistochemistry and PCR. PDHA1 and CDKN2A were obtained as specific cuproptosis-related biomarkers for osteosarcoma after artificial intelligence analysis. PDHA1, CDKN2A and neutrophils were differentially expressed in OS and control groups. PDHA1 and CDKN2A are significantly dysregulated in OS and are able to serve as biomarkers of OS.