A new study has used machine learning algorithms and big data in the medical field to screen for the optimal gene signature in pulmonary arterial hypertension (PAH).

The study, published in the Scottish Medical Journal, screened 564 differentially expressed genes in samples from PAH and controls.

“A comprehensive analysis involving bioinformatics and clinical practice can be used to screen potential genes involved in the pathogenesis of various diseases, which is helpful for the diagnosis and treatment of various diseases,” the authors wrote. “The purpose of this study was to explore potential genetic signals to further understand the pathogenesis of PAH.”


Continue Reading

The research team used least absolute shrinkage and selection operator and support vector machine algorithms in the screening process. In addition, they used the RNA deconvolution tool CIBERSORT to explore the role of various types of immune cells in PAH.

Read more about PAH etiology

The aim was to identify the potential predictive value of various gene expression levels in patients with PAH compared with their expression in healthy controls. In addition, tissue gene expression profiles were used to determine the quantity of cellular components and the degree of infiltration of immune cells in PAH samples. Identifying PAH-specific early immune gene signals and patterns of immune cell infiltration could aid in improving the prognosis for patients with PAH and suggesting new immune-related treatment approaches.

The bioinformatics analyses identified 564 differentially expressed genes in PAH, revealing CALD1 and SLC7A11 to be associated with immune cell activation and to play key roles in PAH. CALD1 and SLC7A11 activity are both likely to participate in the pathogenesis of PAH; however, CALD1 is associated with T cells and SLC7A11 is associated with inflammatory T cells, mast cells, and neutrophils.

The authors strongly suspect that both genes participate in the pathogenesis of PAH and could become potential new targets for immunotherapy in PAH. They caution that in vivo and in vitro studies are still needed to verify their conclusion.

Reference

Jiang C, Jiang W. Lasso algorithm and support vector machine strategy to screen pulmonary arterial hypertension gene diagnostic markers. Scott Med J. Published online October 17, 2022. doi:10.1177/00369330221132158